Shakespeare and the Chicken Egg
Sean Pitman
> It is far from clear that there is any difference between the kind of
> process which generates a chicken from a chicken egg and the kind of
> process which allows William Shakespeare to write "Hamlet".
Oh really? Hmmmmm . . . Do you really think that the chicken egg was
as creative it its forming of the chicken as Shakespeare was in his
forming of Hamlet? For example, a computer can be programmed to do
fantastic things, but it is not creative. Now granted, the terms
"intelligence" and "creative" have not yet been absolutely defined and
maybe they never will be. However, they are defined enough for us to
know that human intelligence can do things that computers and eggs
cannot do. Humans can create new things at high levels of functional
complexity that we never created before and were not preprogrammed to
create automatically. An egg or a computer program cannot create new
things that they were not already programmed to create. A chicken egg
cannot make anything except for a chicken. Shakespeare, on the other
hand, was not preprogrammed to make Hamlet or Macbeth or the Taming of
the Shrew. Though these creations are admittedly not as functionally
complex as a chicken, the process involved in their creation was much
more creative. If Shakespeare had figured out how to make a chicken
without some sort of internal preprogramming, then that would have
been very creative indeed. The fact of the matter is, just because a
computer can do something better or even at a higher level of
complexity than you can do does not make the computer more creative
than you are. Wouldn't you agree that this is a significant
difference between Shakespeare and the chicken egg?
>You are
> merely presuming your conclusion when you say that "no such mindless
> process can give rise to a greater level of complexity... that goes
> very far beyond what its original programming allowed it to do".
This conclusion happens to be my hypothesis. That is what the
scientific method is all about. You observe a given phenomenon and
then make a conclusion/hypothesis to explain this phenomenon. This is
a valid scientific process as long as the hypothesis makes a testable
prediction that can in fact be disproved or "falsified". This is what
I have done. I have predicted that no mindless process will ever be
able to create anything new within a given level of complexity or
beyond in real time. I have drawn this line at several thousand amino
acids working at the same time. So far, the highest level of
functional complexity that has been observed to evolve in real time
requires less than a few hundred amino acids at minimum for that type
of function (i.e., the lactase or nylonase functions). Nothing beyond
such levels of complexity have ever been shown to evolve in real time
and even many life forms seem to be incapable of evolving much of
anything requiring only a few hundred amino acids working at the same
time. For example, many types of bacteria, to include Hall's double
mutant E. coli bacteria, cannot evolve the relatively simple lactase
function in over a million generations of positive selection pressure.
Hall himself referred to these bacteria as having, "limited
evolutionary potential." Now I find that most interesting . . .
> The
> statement itself attempts to confuse the issues by using a term which
> we normally associate with human effort (namefly 'programming') with
> something that seldom does (namely biological development).
Biological development certainly involves pre-established information
systems of extraordinary informational complexity. Without this
information being there fully formed, random organic matter doesn't
turn into much of anything besides amorphous ooze, much less a
chicken. The pre-established information system is vital to the
functional organizational ability of the chicken egg . . . and all
other biological activities. For example, the parts of a flagellum,
if added to solution suddenly or randomly, will not self-assemble. A
very specific order and concentration of part additions is required in
order for the flagellum to form in such a way that its motility
function will be realized. This specific order requires a
pre-established information system and physical apparatus to decode
this information before a motile flagellum can be built. Information
systems at such levels of complexity simply do not self-assemble
without outside input from some higher information system or
intelligence.
<snip>
> This may be what you think the problem is in a nutshell, but it
> unfortunately has no evidence to back it up at all.
What would you call the fact that there is a ladder of complexity
where evolution works very well on the lowest rungs, but less and less
well as it tries to move up the ladder to higher levels of functional
complexity (involving more and more amino acids at minimum)?
For example, very simple functions, such as many forms of antibiotic
resistance, which work by blocking or interfering with other
pre-established functions or interactions, evolve commonly and
rapidly. This is because there are many different ways, involving
only one or two point mutations, to interfere with the
antibiotic-target interaction. So, there are a lot of beneficial
sequences surrounding the starting sequence. Like stepping-stones,
the mindless evolutionary processes of random mutation and natural
selection can quickly cross over toward the move beneficial levels of
more and more efficient antibiotic resistance. However, functions
that require independent action, as is the case with enzymatic
functions, are much more difficult to evolve since there are far fewer
ways for a series of amino acids to achieve a particular enzymatic
function. Still, those enzymatic functions that require fewer amino
acids at minimum in the least specified order are the easiest ones to
evolve. In fact, although there are far fewer examples of novel
enzymatic functions evolving, they are still fairly common - and more
common for shorter enzymes. However, for those functions that require
more and more amino acids, at minimum, working together at the same
time in a fairly constrained manner, evolution becomes exponentially
less and less common. In fact, many life forms, such as many types of
bacteria, simply cannot evolve something like a relatively simple
lactase function, which requires, at minimum, only 400 or 500 amino
acids in a fairly flexible order.
Some in this forum, such as Von Smith and a few others, have suggested
that the ratio of lactase to non-lactase sequences at this level of
complexity is as high as 1 in 1,000 sequences. The problem here is
that Von clearly doesn't understand the power of random walk. If the
ratio were truly this high, only 1,000 mutations would be needed, on
average, to find a lactase sequence in sequence space. An average
bacterial colony would realize such a sequence many its members in
just one or two generations. Evolution at such a high ratio for
success would not only be guaranteed, it would be rapid. The fact is
that E. coli, without the lacZ and ebg genes, do not evolve the
lactase function despite tens of thousands of generations under high
selection pressures, high mutation rates, and very large population
numbers. Other bacteria haven't evolved this function either in over
a million generations of time - and Von thinks that the ratio is 1 in
1000? Please! You've got to be kiddin me!
The problem is that at increasing levels of minimum amino acid
requirements, the ratio of beneficial vs. non-beneficial goes down
dramatically so that not only can no new types of functions be evolved
at higher levels of complexity, they cannot be evolved even within the
same level of complexity. It is like the diagram figured below where
simple beneficial islands are clustered close together, but get
farther and farther apart as one moves up the ladder of complexity
(each dot represents a beneficial sequence in sequence space).
_______________________________________
…………………………………………………..
. . . . . . . . . . . . . . . . . . . .
. . .
. . . . . . . . .
.
. . . .
.
. . .
.
.
.
> Your conclusion of
> an intelligent designer is based upon the improbability of long chains
> of amino acids forming randomly, but that's rather silly and bears no
> resemblance to any modern theory of evolution/abiogenesis/genetics.
Actually that is exactly what the theory of evolution suggests.
Random mutations are supposed to find new beneficial functions, which
can be selected in a positive way by Mother Nature.
> The truth is that long chains of amino acids in very specific sequences
> DO form, form with great frequency.
Actually they do not form with great frequency outside of the
pre-established information system in the DNA of that creature which
codes for their formation.
> If you are to claim that they are
> somehow designed, it is up to you to present evidence that they are
> designed. Merely asserting that some particular model of random formation
> makes them exceptionally unlikely is not sufficient.
Actually it is sufficient. The detection of intelligent activity at
the level of humans or beyond is based on two things: 1) That such
levels of intelligence are capable of producing a given phenomenon,
and 2) that no lesser intelligence or other mindless process is
capable of producing anything even close. Once these two things are
known, intelligent design can be reasonably inferred with a very high
degree of predictive value.
Also, I've done a lot more than just assert than the success of a
random walk will be unsuccessful at a certain level of functional
complexity (involving a few thousands amino acids at minimum), I have
demonstrated a real-life exponential decline in the ability of
mindless evolutionary processes to produce new types of functions at
higher and higher levels of complexity. I have also proposed a very
reasonable explanation for this decline in ability found in the form
of a neutral gap problem. Each additional minimum amino acid
requirement expands the sequence space 20 fold. However, the number
of beneficial sequences only expands a fraction of this amount. This
creates an exponential decrease in the ratio of beneficial vs.
non-beneficial sequences in sequence space. The stepping-stones
become more and more widely separated on average. Very soon the
average gap between beneficial stepping-stones is truly enormous. The
only way for the mindless processes of evolution to bridge this gap is
via random walk alone since nature cannot select between equally
non-beneficial sequences even though they may be different in amino
acid "spelling".
Sean
Deaddog
"Sean Pitman" <seanpi...@naturalselection.0catch.com> wrote in message
news:80d0c26f.03120...@posting.google.com...:
> I have predicted that no mindless process will ever be
> able to create anything new within a given level of complexity or
> beyond in real time. I have drawn this line at several thousand amino
> acids working at the same time.
Well, um, who gives a rat's ass? You can define it to be 'my big floppy
friend the scarecrow' for all I care. A strawman is a strawman is a
strawman. No one claims that complex enzymatic functions arose with great
alacrity from long strings of random protein sequence information. This has
of course been tested: Keefe and Szostak (2001), Nature 410:715 (the enzyme
work has yet to appear, though). High-affinity binders similar to modern
catalysts don't occur frequently (although if I read your argument
correctly, you would have predicted not at all?). Duh. These experiments
are cool but not necessarily relevant to origins, and certainly not relevant
to understanding modern protein evolution, which as we have previously
stated and as you have previously ignored occurs in a wonderfully modular
fashion that allows high order complexity to be achieved.
Similarly, your attempts to squish Hall's most excellent experiments on the
evolutionary potential of cryptic sequences into some sort of diatribe on
design are just ... weird. If you want to tilt at windmills, at least have
the good sense to trade in your wheelbarrow for a horse, your pixie stick
for a lance, and your intellect for ... well, we shouldn't move outside the
realm of possibility.
On the other hand, many people would claim that simple enzymatic functions
that could support relatively limited metabolisms evolved at one point, and
that more complex function arose from duplication and diversification of
these simple functions. There is an abundance of evidence to support this
non-strawman theory of enzyme function and metabolic origins. For example,
simple peptides are more than capable of acting as catalysts. A former
advisor used to chortle that lysine is a kick-ass oxaloacetate
decarboxylase, and it is. And the formation of protein structure, the
precursor to common catalytic function, is also surprisingly easy. I think
one of the best papers in this regard is Kamtekar et al. (1993), Science
262:1680. Basically, even very simple codes based on very simple
physicochemical models can lead to the formation of coherent protein
structures. Follow-up work by the Hecht lab and others has nicely confirmed
this.
Now, would we like to see simple peptide catalysts get incorporated into
simple protein structures, and for their function to thereby improve? Sure.
Would we like to see simple enzymes get together to take on more complex
tasks? You betcha. Do you for a moment believe that either of these is
full-stop outside the realm of realistic probabilities? Sure you do, what
am I thinking (slaps head)! OK, look, set some realistic guidelines,
including, say kcat's or Km's or number / type of chemical transformations
or whatnot; I'll come back and slap your guidelines around a bit, but I'm
sure we can come to an agreement that will allow you to be proved wrong
within a few years, if not immediately. Tell you what, we can even expand
your Universe to include nucleic as well as amino acids, that should allow
us to probe even more realistic origins scenarios.
See, this is how science is done. Hypothesis. Experiment. Ass-kicking.
Take floppy there back to the field for the crows to chew on; he's done.
Non-woof
Tracy Hamilton
"Deaddog" <elling...@yahoo.com> wrote in message
news:bqlrvp$os5$1...@geraldo.cc.utexas.edu...
[snip]
> On the other hand, many people would claim that simple enzymatic functions
> that could support relatively limited metabolisms evolved at one point,
and
> that more complex function arose from duplication and diversification of
> these simple functions. There is an abundance of evidence to support this
> non-strawman theory of enzyme function and metabolic origins. For
example,
> simple peptides are more than capable of acting as catalysts. A former
> advisor used to chortle that lysine is a kick-ass oxaloacetate
> decarboxylase, and it is.
<ID mode>
Lysine is an irreducibly complex peptide (remove just one amino
acid and it loses all function), therefore it is unlikely to have evolved.
If you don't think it is complex, give me its exact wavefunction,
putrescent pup!
</ID mode>
[snip]
Tracy P. Hamilton
r norman
Sorry, but if you want to go into snarky mode, you have to be
punctiliously correct. Lysine is an amino acid, not a peptide. Your
point is still valid -- it is "irreducibly complex". Complex, as you
say, by trying to compute its wavefunction. Irreducible because
removing any single atom from it makes it into something else
entirely. The only way lysine could be produced is "all of a piece",
not from any smaller precursors, according to ID mode thinking.
AR
Sean Pitman wrote:
> >You are
> > merely presuming your conclusion when you say that "no such mindless
> > process can give rise to a greater level of complexity... that goes
> > very far beyond what its original programming allowed it to do".
>
> This conclusion happens to be my hypothesis. That is what the
> scientific method is all about. You observe a given phenomenon and
> then make a conclusion/hypothesis to explain this phenomenon. This is
> a valid scientific process as long as the hypothesis makes a testable
> prediction that can in fact be disproved or "falsified". This is what
> I have done.
Pitman does not understand the scientific process. He has the components
of the process confused, and blended together into a mish mash.
I've pointed this out to him before. And he has always been unwilling
to discuss his mistaken notions about the scientific process.
Gee, isn't that a surprise.
His confusion is one reason why he has a tendency to present many of
his components of the scientific process as pairs of what are actually
separate ideas, e.g., "conclusion/hypothesis". These of course do not have
the same meanings in the scientific process.
He cannot produce the correct definition of a theory or a hypothesis
and he cannot explain the role or either in the scientific process.
Of course he is presuming his conclusion as part of his
"prediction/hypothesis/theory/lunch/dinner/scientific test/...
He has little or no idea of how the scientific process works
or what the terms mean.
Regards
Sean Pitman
> news:80d0c26f.03120...@posting.google.com...:
> > I have predicted that no mindless process will ever be
> > able to create anything new within a given level of complexity or
> > beyond in real time. I have drawn this line at several thousand amino
> > acids working at the same time.
>
> Well, um, who gives a rat's ass? You can define it to be 'my big floppy
> friend the scarecrow' for all I care. A strawman is a strawman is a
> strawman. No one claims that complex enzymatic functions arose with great
> alacrity from long strings of random protein sequence information.
Talk about a strawman! I'm not even saying that complex enzymatic
functions need to arise from random protein sequences. I'm saying
that starting with something that already works evolution will not be
able to evolve anything else that works in a different way at the same
level of complexity or greater beyond the lowest levels of functional
complexity. In other words, if the starting point is at a level of
functional complexity that requires a few thousand amino acids working
together at the same time in a fairly specified sequential order, no
new types of function will evolve at that level or beyond.
This is no strawman. This is exactly what evolutionists claim that
evolution has done and is capable of doing. And yet, evolutionists
have failed to show how such novel functional evolution is remotely
possible. The neutral gaps involved are truly enormous and expand
exponentially with each additional minimum amino acid requirement.
How are the crossing of such gaps achieved via the mindless processes
of random mutation and natural selection regardless of the original
functional sequences that one starts with?
> This has
> of course been tested: Keefe and Szostak (2001), Nature 410:715 (the enzyme
> work has yet to appear, though). High-affinity binders similar to modern
> catalysts don't occur frequently (although if I read your argument
> correctly, you would have predicted not at all?).
Obviously you don't read my argument correctly. Such binding
affinities are not a problem at all since they require few amino acids
and little specificity. The minimum amino acid requirement is quite
small indeed. The level of functional complexity is therefore very
low indeed.
> Duh. These experiments
> are cool but not necessarily relevant to origins, and certainly not relevant
> to understanding modern protein evolution, which as we have previously
> stated and as you have previously ignored occurs in a wonderfully modular
> fashion that allows high order complexity to be achieved.
Again, you repeat these bold statements but give no evidence to
support yourself beyond the historical demonstration of similarities.
You have not even attempted to explain how the gaps between different
kinds of highly complex functions could have been crossed. You just
have this amazing faith that they were crossed via mindless
evolutionary processes alone based on weak historical observations
that there are certain similarities between various functional
systems. You and many other evolutionists fall into the same classic
error of thinking that similarities support the idea of common
evolutionary origin over the idea of common intelligent design.
Similarities support both positions equally well. The only way you
can rule out the idea that intelligence and only intelligence (at the
level of humans or beyond) could have given rise to such levels of
functional complexity is by showing that a mindless process can
actually approach something, anything, within such a level of
functional complexity. This is a falsifiable hypothesis, as
falsifiable as the idea that no cow can jump over my house, and yet
you evolutionists haven't even come close to falsifying this position.
Statistically, it is impossible this side of zillions of years. You
have not overcome this problem in an even remotely convincing way.
You set up your scarecrows, but have little else to offer.
> Similarly, your attempts to squish Hall's most excellent experiments on the
> evolutionary potential of cryptic sequences into some sort of diatribe on
> design are just ... weird.
How so? What Hall did clearly shows the limits of evolutionary
processes. His experiments clearly show that the ratio of certain
functions in sequence space is quite low indeed and that they get
exponentially lower at higher and higher levels of functional
complexity (i.e., more and more amino acids required at minimum).
How is this idea "weird"?
> If you want to tilt at windmills, at least have
> the good sense to trade in your wheelbarrow for a horse, your pixie stick
> for a lance, and your intellect for ... well, we shouldn't move outside the
> realm of possibility.
Nice verbiage . . . but what the heck does it mean? What do you have
as evidence against my position, predictions, and hypothesis besides
meaningless statements like this?
> On the other hand, many people would claim that simple enzymatic functions
> that could support relatively limited metabolisms evolved at one point, and
> that more complex function arose from duplication and diversification of
> these simple functions. There is an abundance of evidence to support this
> non-strawman theory of enzyme function and metabolic origins. For example,
> simple peptides are more than capable of acting as catalysts.
Yes, like the lactase and nylonase enzymes, which requires no more
than 480 or so amino acids in a fairly flexible order at minimum.
Nothing much higher than this level of complexity has ever been shown
to evolve. Evolution simply stalls out at this rather low end of the
spectrum of functional complexity. What else do you have?
> A former
> advisor used to chortle that lysine is a kick-ass oxaloacetate
> decarboxylase, and it is. And the formation of protein structure, the
> precursor to common catalytic function, is also surprisingly easy. I think
> one of the best papers in this regard is Kamtekar et al. (1993), Science
> 262:1680.
How many amino acids at minimum were required? That is the question.
What do you have in regards to this question.
> Basically, even very simple codes based on very simple
> physicochemical models can lead to the formation of coherent protein
> structures. Follow-up work by the Hecht lab and others has nicely confirmed
> this.
How many amino acids required at minimum for the beneficial functions
of these protein sequences?
> Now, would we like to see simple peptide catalysts get incorporated into
> simple protein structures, and for their function to thereby improve?
Actually, this does happen, but the individual functions, the protein
domains, do not improve above the level of functional complexity
requiring a few hundred amino acids working together at the same time.
> Sure.
> Would we like to see simple enzymes get together to take on more complex
> tasks? You betcha.
Oh, I bet you would, but it just doesn't happen now does it?
> Do you for a moment believe that either of these is
> full-stop outside the realm of realistic probabilities? Sure you do, what
> am I thinking (slaps head)!
Keep slapping that head of yours until you find some realistic
explanation or an actual example that proves me wrong.
> OK, look, set some realistic guidelines,
> including, say kcat's or Km's or number / type of chemical transformations
> or whatnot; I'll come back and slap your guidelines around a bit, but I'm
> sure we can come to an agreement that will allow you to be proved wrong
> within a few years, if not immediately.
I have set guidelines. The guidelines that I have set require
functions will minimal amino acid requirements of a few thousand amino
acids working together at the same time in a fairly specified order to
be evolved. Such levels of complexity are very common in all living
things. For example, all bacterial motility systems require several
thousand rather specified amino acids, in the form of several
different proteins, working together at the same time for the function
of motility to be realized in a beneficial way. If you can show any
new function evolving within such a level of functional complexity,
you will have something. Until then, you have nothing but a scarecrow
theory.
> Tell you what, we can even expand
> your Universe to include nucleic as well as amino acids, that should allow
> us to probe even more realistic origins scenarios.
Fine. Amino acid sequences are coded for by nucleotide sequences by a
minimum ratio of 3:1. For each additional amino acid requirement, the
sequence space also increases by a factor of 3. However, the
beneficial nucleotide sequences in sequence space do no increase by a
factor of three at each successive increase in functional complexity.
That is your problem either way you look at it. Each step up the
ladder of complexity (minimum amino acid or nucleic acid requirement)
results in an exponential increase in the size of sequence space
relative to the number of beneficial sequences contained by that
space. Like rapidly separating stepping-stones, these beneficial
sequences are soon so far apart on average that trillions upon
trillions of years of random walk simply are not enough to cross
through all the non-beneficial sequences that separate the beneficial
sequences in sequence space. The problem here is that natural
selection can select, in a positive way, only those sequences that
have some sort of beneficial function. If beneficial sequences are
very far apart in sequence space, requiring more and more mutations to
reach, the random walk involved gets exponentially longer with each
additional step that is required before a new type of beneficial
function is realized.
> See, this is how science is done. Hypothesis. Experiment. Ass-kicking.
Exactly. You have your hypothesis, but no experimental support. So,
for now, it is your ass with the boot in it.
> Take floppy there back to the field for the crows to chew on; he's done.
That's for sure! Your floppy theory of evolution has little left for
those with half a mind to understand the statistical problems with the
theory. Only those who are devoted to it as "more than a theory" have
the religious fortitude to stick by such an ailing theory.
> Non-woof
Woof
Deaddog
"AR" <xy...@pitt.edu> wrote in message news:3FCF6409...@pitt.edu...:
> Pitman does not understand the scientific process. He has the components
> of the process confused, and blended together into a mish mash.
> I've pointed this out to him before. And he has always been unwilling
> to discuss his mistaken notions about the scientific process.
This explanation certainly makes much of the gibberish that passes as a
reply easier to deal with, if not understand.
> He cannot produce the correct definition of a theory or a hypothesis
> and he cannot explain the role or either in the scientific process.
Right, then we should do the job for him:
So, Sean, if we take some thousand or more amino acids that have never
worked together before and show that they can now evolve to work together in
a functional way, that would satisfy your 'test,' correct?
And, to avoid an avalanche of "Pay me $150 you lying weasel" strings, please
specify exactly what you mean so that it can be independently verified
whether or not it has already been demonstrated by science. I think my
definition, above, is pretty nifty: we find some set of proteins (whose sum
is > 1000 amino acids) and that have never worked together before in
biology, and show that by using evolution they can now work together to form
a functional pathway. Isn't that really what you're trying to get at?
Otherwise, well, so far all you've written is a bunch of gibberish that
cannot in any way, shape, or form be reduced to a question, hypothesis,
theory, or even a decent stain under a coasterless glass of lemonade.
Non-woof
howard hershey
Sean Pitman wrote:
> "Deaddog" <elling...@yahoo.com> wrote in message news:<bqlrvp$os5$1...@geraldo.cc.utexas.edu>...
>
>>"Sean Pitman" <seanpi...@naturalselection.0catch.com> wrote in message
>>news:80d0c26f.03120...@posting.google.com...:
>
>
>>>I have predicted that no mindless process will ever be
>>>able to create anything new within a given level of complexity or
>>>beyond in real time. I have drawn this line at several thousand amino
>>>acids working at the same time.
>>
>>Well, um, who gives a rat's ass? You can define it to be 'my big floppy
>>friend the scarecrow' for all I care. A strawman is a strawman is a
>>strawman. No one claims that complex enzymatic functions arose with great
>>alacrity from long strings of random protein sequence information.
>
>
> Talk about a strawman! I'm not even saying that complex enzymatic
> functions need to arise from random protein sequences.
When you say that in order for a new function to arise you must change
"a few thousand amino acids" you are indeed saying that "complex
enzymatic functions need to arise from random protein sequences". At
least to the extent that you are saying anything.
You are *specifically* denying, for example, that a single mutation can
lead two proteins that previously did not interact to interact. You are
*specifically* denying that a single mutation can lead two proteins that
currently interact to interact in a different fashion, perhaps with a
modified function. You are *specifically* claiming that the only way to
generate a two-protein complex that performs a new function is to start
from scratch and go through "several thousand amino acid" changes.
Otherwise, there is no difference between your "lowest levels of
functional complexity" and "higher levels of functional complexity".
You agree that single mutations can alter function at low levels of
functional complexity. What prevents it from happening at high levels?
You claim it is because at high levels, unlike at lower levels, you
need thousands of amino acid changes.
In short, you avoid the strawman "number of amino acids" argument when
you discuss antibiotic resistance and similar systems only to re-invent
it when the system is at a "higher level of complexity". You NEVER
justify your reasoning for this sudden rapid change in number of events
needed to effect functional change.
> I'm saying
> that starting with something that already works evolution will not be
> able to evolve anything else that works in a different way at the same
> level of complexity or greater beyond the lowest levels of functional
> complexity. In other words, if the starting point is at a level of
> functional complexity that requires a few thousand amino acids working
> together at the same time in a fairly specified sequential order, no
> new types of function will evolve at that level or beyond.
I did mention reversion of any of several mutations of bacterial
flagella which produce immotile flagella which still retain protein
export abilities, did I not (although the forward mutation would work as
well)? You agree that bacterial flagella are multiprotein complex
systems involving thousands of amino acids, right? You agree that it
has a motility function, right? Some bacterial fagella also have a
protein export capability, right? You would agree that single mutations
can produce a complex, multiprotein system that only have the second
function. Would you then agree that reversion (also a single mutation)
can produce a new function (motility) from the system with only protein
export utility?
>
> This is no strawman. This is exactly what evolutionists claim that
> evolution has done and is capable of doing.
No it isn't. What scientists propose does not involve thousands of
selectively neutral changes that must all occur (in order?) in order to
produce a selectable effect. It involves duplications and divergence
involving a *few* or even *one* change. It involves changes that allow
multifunctionality for individual proteins (also single or a few
changes). In NO case do molecular evolutionary biologists propose
starting with some quasi-random sequence and randomly walking through
selectively neutral space for thousands of changes in amino acids until
the one keystone change magically produces the only possible selectable
change. Now proteins may indeed undergo thousands of selectively
neutral changes, but these changes are typically selectively irrelevant
in that it doesn't matter whether any or all or an entirely different
set of changes occurred. All that matters is the one or few mutations
of selective relevance.
> And yet, evolutionists
> have failed to show how such novel functional evolution is remotely
> possible. The neutral gaps involved are truly enormous and expand
> exponentially with each additional minimum amino acid requirement.
> How are the crossing of such gaps achieved via the mindless processes
> of random mutation and natural selection regardless of the original
> functional sequences that one starts with?
Because there are no such enormous neutral gaps that *must* be crossed.
You are positing a problem that could only have meaning if you started
with a protein which was essentially (relative to the end protein) a
random sequence with no similarity or active sites with related function
to the end result. Evolutionary biologists *specifically*, by pointing
out the likely precursor proteins, which *always* have sequence
similarity (especially in the functional active sites), disagree that
evolution happens by the strawman "thousands of changes" model you present.
>>This has
>>of course been tested: Keefe and Szostak (2001), Nature 410:715 (the enzyme
>>work has yet to appear, though). High-affinity binders similar to modern
>>catalysts don't occur frequently (although if I read your argument
>>correctly, you would have predicted not at all?).
>
>
> Obviously you don't read my argument correctly. Such binding
> affinities are not a problem at all since they require few amino acids
> and little specificity. The minimum amino acid requirement is quite
> small indeed. The level of functional complexity is therefore very
> low indeed.
Then where do you think the "thousands of amino acid" changes come in?
>>Duh. These experiments
>>are cool but not necessarily relevant to origins, and certainly not relevant
>>to understanding modern protein evolution, which as we have previously
>>stated and as you have previously ignored occurs in a wonderfully modular
>>fashion that allows high order complexity to be achieved.
>
>
> Again, you repeat these bold statements but give no evidence to
> support yourself beyond the historical demonstration of similarities.
> You have not even attempted to explain how the gaps between different
> kinds of highly complex functions could have been crossed.
How big a gap exists between the alpha globin of hemoglobin and the beta
globins of hemoglobin? Why is it impossible to produce the latter by
duplication and divergence from the former, with the irreducible
complexity of hemoglobins in organisms with both being an emergent
property? Where are the "enormous neutral gaps" that prevent the
formation of a 'system' that involves these two proteins interacting
with each other at the same time and place to produce a function?
> You just
> have this amazing faith that they were crossed via mindless
> evolutionary processes alone based on weak historical observations
> that there are certain similarities between various functional
> systems.
The similarities show which small sections of proteins are crucial to
function and which parts are irrelevant to function. It also shows
which parts had to change to change function. These are often small and
require only a few changes.
> You and many other evolutionists fall into the same classic
> error of thinking that similarities support the idea of common
> evolutionary origin over the idea of common intelligent design.
> Similarities support both positions equally well.
Only if the designer chose to produce the type of nested hierarchy that
is also produced by history rather than the type of changes that human
designers would make (involving extensive borrowing and changes only in
the functionally relevant rather than the functionally irrelevant parts
to acheive local adaptive features). But since the designer is your
HYPE (hypothetical posited entity), you can design him/her/it/they any
way you want to do anything you want him/her/it/them to do.
> The only way you
> can rule out the idea that intelligence and only intelligence (at the
> level of humans or beyond) could have given rise to such levels of
> functional complexity is by showing that a mindless process can
> actually approach something, anything, within such a level of
> functional complexity.
And this "level of functional complexity", at its minimum, must be
what....? Two proteins interacting? Three? Proteins that total 600
amino acids? 1200? Do they have to have 20 amino acid sites that
*must* change from some hypothetical precursor protein? 200? 2000?
You never do get around to actually telling us what you mean by "level
of complexity" and how we would recognize it, do you? All you do is say
that for things like antibiotic resistance the complexity is low,
involving as little as a single amino acid change, but for things that
involve 2?, 3? proteins working together, it involves "thousands of
amino acids". Is that "thousands of amino acids" that must change to
reach a new function or just "thousands of amino acids" in the protein?
You never say. Evasion. Evasion. Evasion. [Or, to be kind,
unclarity of thought.]
> This is a falsifiable hypothesis, as
> falsifiable as the idea that no cow can jump over my house, and yet
> you evolutionists haven't even come close to falsifying this position.
> Statistically, it is impossible this side of zillions of years. You
> have not overcome this problem in an even remotely convincing way.
> You set up your scarecrows, but have little else to offer.
Yet you never explicitly say what step cannot be accomplished. All you
say is that it is the step involving "thousands of amino acids" or
"enormous neutral gaps" or some other word salad.
>>Similarly, your attempts to squish Hall's most excellent experiments on the
>>evolutionary potential of cryptic sequences into some sort of diatribe on
>>design are just ... weird.
>
>
> How so? What Hall did clearly shows the limits of evolutionary
> processes. His experiments clearly show that the ratio of certain
> functions in sequence space is quite low indeed and that they get
> exponentially lower at higher and higher levels of functional
> complexity (i.e., more and more amino acids required at minimum).
> How is this idea "weird"?
It is weird because you seem to think that evolution is 'magic' just
like your preferred mechanism (an intelligent something swoops down and
poofs whatever is needed into existence from nothing at all). Evolution
does not claim that one can generate any activity from any random
sequence in five years by the application of a little selective
pressure. It never did. Evolution works by "descent with modification"
not "magic thousands of changes". It always has to start with something
which can reasonably be modified to a new or useful function.
>>If you want to tilt at windmills, at least have
>>the good sense to trade in your wheelbarrow for a horse, your pixie stick
>>for a lance, and your intellect for ... well, we shouldn't move outside the
>>realm of possibility.
>
>
> Nice verbiage . . . but what the heck does it mean? What do you have
> as evidence against my position, predictions, and hypothesis besides
> meaningless statements like this?
Your position is a tilting against a windmill (or a strawman). Your
argument is simply irrelevant.
>>On the other hand, many people would claim that simple enzymatic functions
>>that could support relatively limited metabolisms evolved at one point, and
>>that more complex function arose from duplication and diversification of
>>these simple functions. There is an abundance of evidence to support this
>>non-strawman theory of enzyme function and metabolic origins. For example,
>>simple peptides are more than capable of acting as catalysts.
>
>
> Yes, like the lactase and nylonase enzymes, which requires no more
> than 480 or so amino acids in a fairly flexible order at minimum.
The number of amino acids is irrelevant. The number of mutational
changes needed to reach a selectable function is not. Can you present a
clear argument wrt why you think the total number of amino acids is
relevant and not simply a number presented in order to generate a
strawman argument?
> Nothing much higher than this level of complexity has ever been shown
> to evolve. Evolution simply stalls out at this rather low end of the
> spectrum of functional complexity. What else do you have?
>
>
>>A former
>>advisor used to chortle that lysine is a kick-ass oxaloacetate
>>decarboxylase, and it is. And the formation of protein structure, the
>>precursor to common catalytic function, is also surprisingly easy. I think
>>one of the best papers in this regard is Kamtekar et al. (1993), Science
>>262:1680.
>
>
> How many amino acids at minimum were required? That is the question.
> What do you have in regards to this question.
Why do you think that the number of amino acids in a protein(s) is
relevant to evolution of new functionality? Other, of course, than that
the number is important in a strawman argument that implies that new
functions arise from random starting sequences.
>>Basically, even very simple codes based on very simple
>>physicochemical models can lead to the formation of coherent protein
>>structures. Follow-up work by the Hecht lab and others has nicely confirmed
>>this.
>
>
> How many amino acids required at minimum for the beneficial functions
> of these protein sequences?
Why do you think that the number of amino acids in a protein(s) is
relevant to evolution of new functionality? Other, of course, than that
the number is important in a strawman argument that implies that new
functions arise from random starting sequences.
>>Now, would we like to see simple peptide catalysts get incorporated into
>>simple protein structures, and for their function to thereby improve?
>
>
> Actually, this does happen, but the individual functions, the protein
> domains, do not improve above the level of functional complexity
> requiring a few hundred amino acids working together at the same time.
Why do you think that the number of amino acids in a protein domain is
relevant to evolution of new functionality? Other, of course, than that
the number is important in a strawman argument that implies that new
functions arise from random starting sequences.
>> Sure.
>>Would we like to see simple enzymes get together to take on more complex
>>tasks? You betcha.
>
>
> Oh, I bet you would, but it just doesn't happen now does it?
>
>
>>Do you for a moment believe that either of these is
>>full-stop outside the realm of realistic probabilities? Sure you do, what
>>am I thinking (slaps head)!
>
>
> Keep slapping that head of yours until you find some realistic
> explanation or an actual example that proves me wrong.
>
>
>> OK, look, set some realistic guidelines,
>>including, say kcat's or Km's or number / type of chemical transformations
>>or whatnot; I'll come back and slap your guidelines around a bit, but I'm
>>sure we can come to an agreement that will allow you to be proved wrong
>>within a few years, if not immediately.
>
>
> I have set guidelines. The guidelines that I have set require
> functions will minimal amino acid requirements of a few thousand amino
> acids working together at the same time in a fairly specified order to
> be evolved.
There are proteins or heterodimers or larger complexes that are several
thousand amino acids long, but I know of none that *require* several
thousand *specific* (or invariant) amino acids for function. Do you?
All the ones I know of are much more flexible wrt sequence than that.
Perhaps the five histones that make up the nucleosome comes closest, but
histones are pretty small (about 740 amino acids adding up all five) and
I doubt that most of the amino acids are absolutely specified (even
though histones have about the most strongly conserved sequences in all
eucaryotes)? Indeed, Michael Behe claimed that one (count 'em, one)
amino acid was always invariant in one of the histones -- I forget which
one -- (but in the same year, 1990, but after he published, several
species of ciliate were found to have a variant amino acid at that
site). Of course, the change was a conservative change, but it was a
change. In all other proteins the number of "invariant" amino acids and
conserved (functionaly relevant) sequence is much smaller.
> Such levels of complexity are very common in all living
> things. For example, all bacterial motility systems require several
> thousand rather specified amino acids, in the form of several
> different proteins, working together at the same time for the function
> of motility to be realized in a beneficial way.
But that does not mean that one has to change several thousand amino
acids in order to generate a bacterial 'mobility' system. One can, as I
point out, change a specific single amino acid and convert a "protein
export" system into a system which is both a "protein export" and a
"mobility" system. Now, admittedly this is a simple reversion so that
the "protein export" system was, in the historical past, a system with a
different function. But as evidence that one does not need "thousands
of amino acid" changes to generate new or altered function, regardless
of how many amino acids or proteins are in the system, I think it makes
its point: The amount of change needed to alter function is not related
to the complexity of the system.
> If you can show any
> new function evolving within such a level of functional complexity,
> you will have something. Until then, you have nothing but a scarecrow
> theory.
If you can present me with a system that could not, by any known
mechanism, including duplication and divergence and chimeric duplication
have evolved into a current without changing "thousands of amino acids",
you would be able to convince me. Until you can present such a system
that requires "thousands of amino acids" to change (i.e., the equivalent
of making new proteins from random sequences), I will regard that as the
classic creationist strawman misunderstanding of how these systems evolve.
>>Tell you what, we can even expand
>>your Universe to include nucleic as well as amino acids, that should allow
>>us to probe even more realistic origins scenarios.
>
>
> Fine. Amino acid sequences are coded for by nucleotide sequences by a
> minimum ratio of 3:1. For each additional amino acid requirement, the
> sequence space also increases by a factor of 3. However, the
> beneficial nucleotide sequences in sequence space do no increase by a
> factor of three at each successive increase in functional complexity.
> That is your problem either way you look at it. Each step up the
> ladder of complexity (minimum amino acid or nucleic acid requirement)
> results in an exponential increase in the size of sequence space
> relative to the number of beneficial sequences contained by that
> space. Like rapidly separating stepping-stones, these beneficial
> sequences are soon so far apart on average that trillions upon
> trillions of years of random walk simply are not enough to cross
> through all the non-beneficial sequences that separate the beneficial
> sequences in sequence space. The problem here is that natural
> selection can select, in a positive way, only those sequences that
> have some sort of beneficial function. If beneficial sequences are
> very far apart in sequence space, requiring more and more mutations to
> reach, the random walk involved gets exponentially longer with each
> additional step that is required before a new type of beneficial
> function is realized.
And positing that beneficial sequences are far apart in sequence space,
requiring more and more mutations to reach does not make it true nor
does it make it the way evolution works (which is by modification of
pre-existing sequences that have function).
>>See, this is how science is done. Hypothesis. Experiment. Ass-kicking.
>
>
> Exactly. You have your hypothesis, but no experimental support. So,
> for now, it is your ass with the boot in it.
Why should I present evidence *for* what I regard as an irrelevant
strawman model of evolution?
>>Take floppy there back to the field for the crows to chew on; he's done.
>
>
> That's for sure! Your floppy theory of evolution has little left for
> those with half a mind to understand the statistical problems with the
> theory. Only those who are devoted to it as "more than a theory" have
> the religious fortitude to stick by such an ailing theory.
I agree that your understanding of statistical problems took half a
mind. I would suggest the lower half.
>
>
>>Non-woof
>
>
> Woof
>
> Sean
> www.naturalselection.0catch.com
>
Mark Isaak
seanpi...@naturalselection.0catch.com (Sean Pitman) wrote:
>Talk about a strawman! I'm not even saying that complex enzymatic
>functions need to arise from random protein sequences. I'm saying
>that starting with something that already works evolution will not be
>able to evolve anything else that works in a different way at the same
>level of complexity or greater beyond the lowest levels of functional
>complexity. In other words, if the starting point is at a level of
>functional complexity that requires a few thousand amino acids working
>together at the same time in a fairly specified sequential order, no
>new types of function will evolve at that level or beyond.
If I understand you correctly, you are saying that it is impossible
for evolution to produce a new functional gene without eliminating at
least one other gene. Is that what you are saying?
--
Mark Isaak at...@earthlink.net
"Voice or no voice, the people can always be brought to the bidding of
the leaders. That is easy. All you have to do is tell them they are
being attacked, and denounce the pacifists for lack of patriotism and
exposing the country to danger." -- Hermann Goering
sweetnes...@yahoo.com
Talk about someone who runs away from evidence, just to post the same,
already disproved nonsense all over again.
>I'm saying that starting with something that
>already works evolution will not be able to
>evolve anything else that works in a different
>way at the same level of complexity or greater
>beyond the lowest levels of functional complexity.
Yes, we know. And it has been explained to you why you are wrong. Just
minutes ago, I did it in your "Mindless Creativity" thread. A few days
ago I did it in the thread on your theory.
>And yet, evolutionists have failed to show how
>such novel functional evolution is remotely possible.
Failed to do the impossivble and force a process that happens over
millenia to happen in the lab within a century.
>The neutral gaps involved are truly enormous and
>expand exponentially with each additional minimum
>amino acid requirement.
Not true. And minimum amino acid requirement is something you
invented.
See the above referenced threads for detailed refutation.
>How are the crossing of such gaps achieved via the
>mindless processes of random mutation and natural
>selection regardless of the original functional
>sequences that one starts with?
I have explained it to you. More then once.
>Such binding affinities are not a problem at
>all since they require few amino acids and
>little specificity. The minimum amino acid
>requirement is quite small indeed. The level
>of functional complexity is therefore very
>low indeed.
And from such small changes, large changes follow.
>Again, you repeat these bold statements but
>give no evidence to support yourself beyond
>the historical demonstration of similarities.
We are observing the processes currently. All that is required is that
those same processes happened in the past. If you have proof that the
laws of the universe were different, and then changed in the last few
centuries, please say so, and provide the evidence.
>You have not even attempted to explain how the gaps
>between different kinds of highly complex functions
>could have been crossed.
Yes we did. I did it myself, in quite a bit of detail, in this
newsgroup.
>You and many other evolutionists fall into the same
>classic error of thinking that similarities support
>the idea of common evolutionary origin over the idea
>of common intelligent design.
Because of its superiority. I'm never sure which of my messages will
make it to the newsgroup first, so I'll quote the last section of my
answer to your "Mindless Creativity" thread (which is, as all other
your threads are, just another repetition):
<start quote>
No faith is needed for evolution. But evolution applies from the first
roughly functioning cell onwards; you might be talking about
abiogenesis, so let's analyze that for a moment. How did the first
cell come about?
This is something that is quite unknown, and entirely theoretical. So,
yes, unlike the case of evolution, one cannot make one's mind based
entirely on evidence. What do we do?
You might propose that, since we don't know, it is equally rational to
simply say that there was an Intelligent Designer who designed the
first cell, as it is to assume that the mindless processes did it. You
might say that it takes faith in both cases, so there is no real
difference.
If you did so (and I don't know if you would, but I have a general
idea), you would be wrong. There is a simple reason to assume that the
first cell arose by mindless processes rather then by intelligent
design. The reason is evidence of existence.
We *know*, without doubt, that mindless processes exist. Everything
that happens, happens according to the laws of the universe. We have
ascertained without much doubt that the same laws (without any large,
visible changes) applied for quite a few billions of years. We know
that those laws were around when the life first began. Therefore, we
have good reason to think that the laws of the universe were involved
in the begining of life.
As for Intelligent Designer...we have no idea if a supernatural entity
such as he exists or not. We see no evidence of his existence, in the
present or in the past. He is entirely in the domain of imagination,
and as such we can imagine anything about him, and therefore measure
nothing. Not only have we no reason to assume his existence, but
taking him as an "option" would literally mean end of any further
research - for if can you learn anything about an invisible,
intanglible being that never does anything, how can you learn anything
about its works?
How did a star form? If we assume it formed by mindless processes, we
can use our knowledge of the laws of physics, and see if we can
discover a solution. After many years of work, we may produce a
theory, compare it to the reality, and see if it stands up to
scrutiny. If we assume that a Designer made it, that is it. We give
up, and say "goddidit", the end. We cannot know anything further.
Through most of the Dark Ages, people took the second approach;
whatever it is, God made it to be that way. This is, in fact, the
reason why that period of time is called "The Dark Ages". Modern
science uses the first approach, and it got as quite a bit further.
So, in case of abiogenesis, I will look at both options: accept a
Designer without *any* reason to do so, plop on a chair, and say
"done". Or assume that the processes that did EVERYTHING ELSE around
me also produced life, and try to figure out how they did it.
Mindless processes, Intelligent Designer, whichever it is, gave us
free will to make our own choices. Each to their own, as they say.
<end quote>
So, even if we ignore the facts (as you do), and claim that there is
no reason to favor one side over the other, there is this reason
above. It is a very good reason, IMO.
>Statistically, it is impossible this side of zillions of years.
Untrue, see referenced posts.
>You have not overcome this problem in an even remotely
>convincing way.
It is impossible to convince someone who refuses to look at the
evidence.
>You set up your scarecrows, but have little else to offer.
Hundreds of thousands of professional scientists do so. Interesting.
>How so? What Hall did clearly shows the limits of evolutionary
>processes. His experiments clearly show that the ratio of certain
>functions in sequence space is quite low indeed and that they get
>exponentially lower at higher and higher levels of functional
>complexity (i.e., more and more amino acids required at minimum).
>How is this idea "weird"?
Because he did nothing of the kind. He showed how functions can
quickly adapt to new conditions, and that a function can arise from a
different one. He also showed that, as expected, evolution works very
slow when there is no basis from which a particular function can
quickly evolve. There was nothing in his findings that contradicted
evolutionary theory in any way, shape or form.
Again, see my answer to you in "Mindless Creativity" thread. To avoid
repetition, I will refer to the previous sentence as "Statement A".
>Yes, like the lactase and nylonase enzymes, which
>requires no more than 480 or so amino acids in a
>fairly flexible order at minimum.
Bull. Statement A.
>Nothing much higher than this level of complexity
>has ever been shown to evolve. Evolution simply
>stalls out at this rather low end of the spectrum
>of functional complexity. What else do you have?
Statement A.
>How many amino acids at minimum were required?
>That is the question.
For which function? There are many functions where one will suffice.
Two, three, five, ten, fifteen...there are examples of all sorts of
short polypeptides performing various functions.
And, again, Statement A.
>What do you have in regards to this question.
Statement A. Plus all the things mentioned in the post within "Sean
Pitman's Theory" thread. Plus about ten-thousandfold as much evidence
in molecular evolution papers published in last fourty years.
>How many amino acids required at minimum for the
>beneficial functions of these protein sequences?
The number of amino acids does not matter. Statement A.
>Actually, this does happen, but the individual functions,
>the protein domains, do not improve above the level of
>functional complexity requiring a few hundred amino acids
>working together at the same time.
But they recombine. They adapt and optimize without "searching"
through every possible combination of residues. Proteines also change
functions, or can be pushed from one function to another by selective
pressures (see the example I gave in the first post in "Sean Pitman's
Theory" thread). Statement A - also some additional details.
>Oh, I bet you would, but it just doesn't happen now does it?
I gave you examples of that happening. Observed. In lab.
>Keep slapping that head of yours until you find
>some realistic explanation or an actual example
>that proves me wrong.
I gave you examples. You ran away, waited for a week, then started
several new threads as if I said nothing.
>I have set guidelines. The guidelines that I have
>set require functions will minimal amino acid
>requirements of a few thousand amino acids working
>together at the same time in a fairly specified order
>to be evolved.
Considering how often we see developments of new, specific enzymes
(quite often), how often we see enzymes that used to work separately
now pairing together to do something else (occasionaly), and how often
we see entire existing system co-opted to perform novel functions
(rarely), it is statistically quite impossible that such a system you
describe won't develop in lab given just a few thousand years of
evolution. Also, Statement A.
>For example, all bacterial motility systems require
>several thousand rather specified amino acids, in
>the form of several different proteins, working
>together at the same time for the function of motility
>to be realized in a beneficial way.
Already covered.
>Until then, you have nothing but a scarecrow theory.
The theory fits all the relevant data, and can be used to make
exceedingly precise predictions. Do you have a theory that will fit
the data better and/or allows for better predictions?
>Fine. Amino acid sequences are coded for by nucleotide
>sequences by a minimum ratio of 3:1. For each additional
>amino acid requirement, the sequence space also increases
>by a factor of 3.
No, it does not. Run "genetic code degeneracy" through a search engine
for explanation why. I give up on getting you to open a book in
biochemistry (and I now utterly disbelieve that you ever took it with
any degree of success; if you passed biochemistry, and you can state
something like this above, your professor should resign immediately).
>The problem here is
...that nothing of what you say has anything to do with reality.
Statement A, plus the referenced thread on your theory.
>Exactly. You have your hypothesis, but no experimental
>support. So, for now, it is your ass with the boot in it.
Tons of experimental support. Plus confirmations from other branches
of science (geology, physics...). But I get tired. Tell me, where is a
singe bit of proof for ID theory? I mean, even if evolution was all
wrong, bogus, product of masonic conspiracy and whatnot, why would I
believe your half-assed bullshit about "Intelligent Designers", rather
then just say "I don't know"?
>Your floppy theory of evolution has little left for
>those with half a mind to understand the statistical
>problems with the theory.
Indeed, the world is fortunate to have people with half a mind, such
as yourself, who can show us all The Truth of Creation. Should I bow
before your magnificence?
Hm. Nah. I'll see what you have to say to the Statement A, and to the
evidence I gave in the "Sean Pitman's Theory" thread (which I should
have named Statement B at the beginning, and save myself quite a bit
of typing).
M.
MEC
Dear Expired Canine
I have been following this discussion. Sort of. And I was wondering
would Bcr/Abl or Ret/PTC chimeric proteins qualify as an example to
the good Dr. Pitman? Both are the results of recombination (chromosome
translocation) and both result in the induction of novel signaling
pathways, among other novel functions. Both fusion genes cause cancer
in humans (chronic myeloid leukemia in the case of Bcr/Abl and
medullary thyroid carcinoma for Ret/PTC). Bcr/Abl is >1500 aa long,
Ret/PTC is > than 1200 aa, depending on isotype of Ret.
Wot you think?
Deaddog
"MEC" <unre...@hotmail.com> wrote in message
news:c240c53.03120...@posting.google.com...
I think that anything that you reveal to be truth in advance of Pitman
agreeing to some set of rules is amazingly not going to be what he means.
Indeed, it seems from the other posters who have commented that this is far
from the first time someone has tried to just put things into a scientific
context for Pitman, only to be rebuffed by garbled, meandering statements
that could not be re-formulated as hypotheses even if one wanted to.
But it's a kick-ass example, and again shows how the evolution of novel
function via modular regulatory domains is quite possible.
Actually, it brings up the rather interesting question: is cancer good for
us, in an evolutionary sense? That is, an Intelligent Designer would
supposedly have tied down all those loose regulatory ends, flopping in the
evolutionary breeze, and thus would not have allowed us to get lumps the
size of pumpkins on our prostates or elsewhere. Ahem. However, this does
happen, because, well, evolution happens (in this case at the cellular
level). These Bad Things can be seen as a loss of organismal fitness, and
over time unsurprisingly result in the evolution of corresponding Good
Things to fix the Bad Things (think tumor suppressor genes).
Does the more complex network that results from selections for fitness
relative to oncogenesis (and other regulatory deficiencies) itself have
greater potential to evolve? Are things like our big ol' brains in part the
product of many years of evolving around regulatory deficiencies? "I
suffered from Lesch-Nyhan syndrome and all I got was this Broca's region!"
Nah. Too weird and speculative and stupid. For now.
Non-woof
Ian Musgrave & Peta O'Donohue
Address altered to avoid spam, delete RemoveInsert
On Fri, 5 Dec 2003 17:37:07 +0000 (UTC), unre...@hotmail.com (MEC)
wrote:
>"Deaddog" <elling...@yahoo.com> wrote in message news:<bqnvet$duf$1...@geraldo.cc.utexas.edu>...
[sean wrote in another post, not in this specific position in this
post
> > I have predicted that no mindless process will ever be
> > able to create anything new within a given level of complexity or
> > beyond in real time. I have drawn this line at several thousand amino
> > acids working at the same time.
just including it for orientation]
>> So, Sean, if we take some thousand or more amino acids that have never
>> worked together before and show that they can now evolve to work together in
>> a functional way, that would satisfy your 'test,' correct?
>>
>> And, to avoid an avalanche of "Pay me $150 you lying weasel" strings, please
>> specify exactly what you mean so that it can be independently verified
>> whether or not it has already been demonstrated by science. I think my
>> definition, above, is pretty nifty: we find some set of proteins (whose sum
>> is > 1000 amino acids) and that have never worked together before in
>> biology, and show that by using evolution they can now work together to form
>> a functional pathway. Isn't that really what you're trying to get at?
>>
>> Otherwise, well, so far all you've written is a bunch of gibberish that
>> cannot in any way, shape, or form be reduced to a question, hypothesis,
>> theory, or even a decent stain under a coasterless glass of lemonade.
>>
>> Non-woof
>
>
>Dear Expired Canine
>
>I have been following this discussion. Sort of. And I was wondering
>would Bcr/Abl or Ret/PTC chimeric proteins qualify as an example to
>the good Dr. Pitman? Both are the results of recombination (chromosome
>translocation) and both result in the induction of novel signaling
>pathways, among other novel functions. Both fusion genes cause cancer
>in humans (chronic myeloid leukemia in the case of Bcr/Abl and
>medullary thyroid carcinoma for Ret/PTC). Bcr/Abl is >1500 aa long,
>Ret/PTC is > than 1200 aa, depending on isotype of Ret.
You beat me to it. These examples are great as
1) A functional protein is created that does not previously exist, eg
BCR-ABL is a tyrosine kinase, and a nice one at that.
2) The proteins have interesting functional features not present in
the original proteins, BCR-ABL has different targeting and substrates
to ABL (we don't know what BCR does yet, it is probably not an enzyme)
3) They are definitely over 1000 amino acids (depending on isotype,
some fusions make short versions, eg there is a "mini" BCR-ABL that is
a mere 800 aa long).
4) They illustrate an important process in protein evolution. Not all
new proteins are generated by single amino acid substitutions, fusions
play a very important role (as does in-gene duplications), an example
is in a recent experiment where new, more efficient yeast hexose
transporters were generated by gene fusion (all done by mindless
processes, the experimenters just stuck the yeast in low hexose medium
and mutation (duplication-fusion) and natural selection did the rest).
Nonetheless, despite clear experimental evidence of generation of
functional enzymes (BCR-ABL tyrosine kinase) with clearly different
specificities (cytoplasmic rather than nuclear targeting) which are
larger than 1,000 aa long, I'm sure these will not be enough for Dr.
Pitman. The key is Dr' Pitman's statement:
"given level of complexity"
He will simply claim that BCR-ABL isn't complex enough. What it does
do is demolish his claim that "neutral gaps" get bigger as proteins
get larger.
When he says
> > I have drawn this line at several thousand amino
> > acids working at the same time.
It is not clear why. As a thought experiment, lets take a
"sufficiently complex" 1,800 aa protein, duplicate it, and fuse the
two together via their respective c-terminal and n-terminal domains,
you then have a very large, sufficiently complex protein working very
nicely (c-n terminal fusions happen all the time, in the lab and in
nature, without compromising the function of the proteins) well over
Dr. Pitman's limit. As another experiment, add a Green-fluorescent
protein domain to the C-terminal end of any large, complex but below
threshold protein, you will now have a large complex protein above Dr.
Pitman's limit that glows green as well.
The irony is that complex systems like the flagella, are not large
proteins, but non-cobvalent assemblies of lots of small-moderate sized
proteins, many of which are clear duplicates of others in the
assembly, so his whole "greater than 1,000 aa" argument does not apply
to them.
Cheers! Ian (posting rarely due to the arrival of Andrew Thomas)
=====================================================
Ian Musgrave Peta O'Donohue,Jack Francis and Michael James Musgrave
reyn...@werple.mira.net.au http://werple.mira.net.au/~reynella/
Southern Sky Watch http://www.abc.net.au/science/space/default.htm
Giant Sloth
As a lurker who admires your ID posts as at least sounding
intelligent, I'm wondering why you leave so many threads hanging, so
many arguments unrefuted in one place and then start a new thread
repeating the same apparently refuted arguments. Admittedly, you're
greatly outnumbered on TO, but could you at least pick one person to
respond to all their arguments? I would nominate
sweetnes...@yahoo.com since he's got unanswered arguments in two
threads and seems to have a good knowledge of the science involved.
He seems very willing to point out your apparent scientific errors.
Ideally, there should be a formal debate, I (and I assume others)
would pay close attention to that one.
RobinGoodfellow
> Actually, it brings up the rather interesting question: is cancer good for
> us, in an evolutionary sense? That is, an Intelligent Designer would
> supposedly have tied down all those loose regulatory ends, flopping in the
> evolutionary breeze, and thus would not have allowed us to get lumps the
> size of pumpkins on our prostates or elsewhere. Ahem. However, this does
> happen, because, well, evolution happens (in this case at the cellular
> level). These Bad Things can be seen as a loss of organismal fitness, and
> over time unsurprisingly result in the evolution of corresponding Good
> Things to fix the Bad Things (think tumor suppressor genes).
>
> Does the more complex network that results from selections for fitness
> relative to oncogenesis (and other regulatory deficiencies) itself have
> greater potential to evolve? Are things like our big ol' brains in part the
> product of many years of evolving around regulatory deficiencies? "I
> suffered from Lesch-Nyhan syndrome and all I got was this Broca's region!"
>
> Nah. Too weird and speculative and stupid. For now.
>
> Non-woof
Weird, yes. Speculative, maybe. Stupid - not necessarily. Check
out:
FW2.2: a quantitative trait locus key to the evolution of tomato fruit
size.
Frary A, Nesbitt, et. al.
Science. 2000 Jul 7;289(5476):85-8.
Quoth the abstract:
Domestication of many plants has correlated with dramatic increases in
fruit size. In tomato, one quantitative trait locus (QTL), fw2.2, was
responsible for a large step in this process .... The cause of the
QTL effect is a single gene, ORFX, that is expressed early in floral
development, controls carpel cell number, and has a sequence
suggesting structural similarity to the human oncogene c-H-ras p21.
OK, granted that a tomato growth spurt is not exactly the same thing
as swelling of brains to human proportions, but it would seem that
oncogenesis can play a positive role in the evolution of at least some
species. (Well, positive for us, anyway - I like my tomatoes big and
juicy.)
Sean Pitman
Sean Pitman
> > Talk about a strawman!
>
> Talk about someone who runs away from evidence, just to post the same,
> already disproved nonsense all over again.
Here is the link to my reply to most of this post in another thread
started by sweetnes (previous link posted wasn't correct):
Sarah Berel-Harrop
"Deaddog" <elling...@yahoo.com> wrote in message
news:bqqva6$cbh$1...@geraldo.cc.utexas.edu...
> Actually, it brings up the rather interesting question: is cancer good
for
> us, in an evolutionary sense? That is, an Intelligent Designer would
> supposedly have tied down all those loose regulatory ends, flopping in the
> evolutionary breeze, and thus would not have allowed us to get lumps the
> size of pumpkins on our prostates or elsewhere. Ahem. However, this does
> happen, because, well, evolution happens (in this case at the cellular
> level). These Bad Things can be seen as a loss of organismal fitness, and
> over time unsurprisingly result in the evolution of corresponding Good
> Things to fix the Bad Things (think tumor suppressor genes).
>
> Does the more complex network that results from selections for fitness
> relative to oncogenesis (and other regulatory deficiencies) itself have
> greater potential to evolve? Are things like our big ol' brains in part
the
> product of many years of evolving around regulatory deficiencies? "I
> suffered from Lesch-Nyhan syndrome and all I got was this Broca's region!"
>
> Nah. Too weird and speculative and stupid. For now.
well, the argument from bad design is fun, i guess,
however it has several deficiencies.
1) it's a strawman. few ID'rs claim the paleyesque
'best of all possible worlds' design scenario. ray
bohlin has a reference on his website about the
'whimsy' of the Creator. this doesn't really count,
because nowhere do i find where he says the
designer's product is perfect from the human point
of view. so now you are off the track with, 'well,
you are misrepresenting my position.' while you
spend time in that argument, you lose the civilians.
2) it's a sideshow. you now have opened up the
opportunity to discuss whether the design is indeed
defective. two problems -- first, you may be perceived
as taking the argument from design seriously. this is
a Very Bad Thing, imv, because it provides a gloss,
however small, of legitimacy. second, you get involved
in the esoteric details of the design scenario. again,
you have confused and lost the civilians.
3) it can be considered offensive. if you do not clearly
differentiated between the theological proposition of
generic design and beauty in nature, as understood by
mainline christians, and Intelligent Design, the
proposition that we may observe God's fingerprints in
this or that phenomenum. now someone who does not
understand that you are talking about Intelligent Design
and not the design of Creation thinks you are slagging
their religion. you have alienated and lost the civilians.
4) this is not a new argument. this is an extension of
the argument from suffering as a dis-proof of God to
biological concepts. this argument is so common that
there are simply reams of apologetics designed to rebut
the argument. now you are arguing outside of your
own area of expertise, and arguing about religion (not
science) in an apologetic engagement, and you have
again lost the civilians who don't find you credible on
this issue (not least because you seem unaware of the
reams and reams of apologetics that they have read,
are aware of, and maybe accept).
in short, you're a jet in shark's territory. to the
extent that you bring up the pattern of evolution to
compensate for the Bad Things That Happen,
and demonstrate the causal link, that is a productive
line of thought. however, again, a bit esoteric,
and difficult to explain why wouldn't the designer
have done it that way as well, to someone that
already has a commitment to generic design (not
necessarily Intelligent Design).
--
I'm standing on the outside of your shelter,
Lookin' in,
While the bombs are here a-droppin'
everywhere ...
Have I ever told you that I care?
(Shel Silverstein)
>
>
Sean Pitman
>
> Dear Expired Canine
>
> I have been following this discussion. Sort of. And I was wondering
> would Bcr/Abl or Ret/PTC chimeric proteins qualify as an example to
> the good Dr. Pitman? Both are the results of recombination (chromosome
> translocation) and both result in the induction of novel signaling
> pathways, among other novel functions. Both fusion genes cause cancer
> in humans (chronic myeloid leukemia in the case of Bcr/Abl and
> medullary thyroid carcinoma for Ret/PTC). Bcr/Abl is >1500 aa long,
> Ret/PTC is > than 1200 aa, depending on isotype of Ret.
>
> Wot you think?
Mutations such as the BCR/ABL translocation between chromosomes 22 and
9 are very interesting. Bcr-Abl is a chimeric protein usually
consisting of the amino terminal 927 amino acids (p210) or the amino
terminal 426 amino acids (p190) of Bcr (breakpoint cluster region)
fused to the second exon of the tyrosine kinase Abl (there are also a
few other potential though less common breakpoints in the Bcr region
and even in the Abl region).
The normal abl kinase has the function of modifying proteins by adding
phosphate to the amino acid tyrosine. This protein is a tightly
regulated tyrosine kinase that is predominantly nuclear in
hematopoietic cells. It has 1130-1143 amino acids and 4 domains.
Phosphorylation of tyrosines is one method by which proteins are
switched from inactive states to active states. Amino acids 381-409
are called the "activation loop". This region is highly conserved in
the family of tyrosine kinase proteins.
The normal bcr protein is also a cytoplasmic kinase with
serine/threonine kinase activity. It is composed of 1271 amino acids
and has 5 domains.
The new fusion protein bcr-abl has an increased kinase function. The
altered kinase function of the new bcr-abl kinase controls the
transformation of a normal blood cell into a cancer cell. The various
known bcr-abl isoforms usually have the same c-abl sequence structure
(with loss of the first abl exon), but may differ in their n-terminal
truncated bcr sequences. This usually involves around 426 or 927
amino acids from bcr (but may be as few as 60 or 70aa) added to the
remaining 900 or so abl amino acids for a total of between 1000 to
1800 amino acids.
In order to understand the function of the chimeric bcr-abl protein it
is helpful to understand how the normal abl tyrosine kinase function
works. Normally a 3BP1 binding protein binds to the normal abl
protein on its SH3 domain. This prevents the activation of the SH1
domain, which has a self-phosphorylable tyrosine. However, the
bcr-able chimeric protein, the first n-terminal exon coded region of
bcr binds to the SH2 domain and hides the SH3 domain. As a
consequence, the 3BP1 binding protein cannot bind to the SH3 domain
and so the SH1 domain remains active on the abl protein. This active
domain activates the RAS signal transduction pathway as well as the
MYC pathway, both of which are involved with oncogenesis.
So, what is happening here is that the normal function of one of the
domains of the abl protein is being enhanced by a block to a domain
that normally binds an inhibitor protein. It is relatively easy to
block a function and therefore cause whatever effect that might
follow. In fact, there are many different ways to enhance the kinase
function of the abl protein and the minimum amino acid requirement to
do so is not that great (probably significantly less than 60aa). The
addition of various bcr sequences ranging from 60 or so amino acids to
over 900aa is not the only way either. Several other translocations
give rise to chimeric proteins that deregulate the kinase function.
Of course, these types of mutations do not really creation a new
function at all, but simply deregulate a pre-established function.
The tyrosine kinase function of the abl protein was already there,
preformed, before its function became enhanced. A relatively small
piece of the bcr protein came along and got stuck to it in such a way
that it blocked part of the abl protein's normal function, thus
deregulating it and causing a marked increase in kinase activity and
ultimate deregulation of several different growth related pathways in
the cell. Notice also that such deregulation results in a selectably
non-beneficial result (i.e., cancer). In fact, a significant number of
cancers are the result of protein kinase deregulation.
In short, the bcr-abl chimeric protein is not a new type of function
nor is it a beneficial enhancement of an old function. It in no way
defeats my claim that novel beneficial protein functions above a few
thousand amino acids cannot evolve via mindless evolutionary
processes.
Sean
www.naturalselection.0catch.com
Online References:
http://www.infobiogen.fr/services/chromcancer/Genes/ABL.html
http://www.bloodjournal.org/cgi/content/full/100/3/1092
http://www.rpc.msoe.edu/cbm2/ablopen.html
http://www.sigmaaldrich.com/sigma/rbi-handbook/sg_ls_cs_rbibook_tyrosine.pdf
http://www.infobiogen.fr/services/chromcancer/Anomalies/t0922ANL.html
http://www.bloodjournal.org/cgi/content/full/99/8/2957
http://www.infobiogen.fr/services/chromcancer/Genes/ABL.html
http://www.nature.com/cgi-taf/DynaPage.taf?file=/onc/journal/v15/n14/abs/1201342a.html&dynoptions=doi1070713813
Ian Musgrave & Peta O'Donohue
Address altered to avoid spam, delete RemoveInsert
On Sat, 6 Dec 2003 15:34:00 +0000 (UTC),
seanpi...@naturalselection.0catch.com (Sean Pitman) wrote:
>unre...@hotmail.com (MEC) wrote in message news:<c240c53.03120...@posting.google.com>...
>> Dear Expired Canine
>>
>> I have been following this discussion. Sort of. And I was wondering
>> would Bcr/Abl or Ret/PTC chimeric proteins qualify as an example to
>> the good Dr. Pitman? Both are the results of recombination (chromosome
>> translocation) and both result in the induction of novel signaling
>> pathways, among other novel functions. Both fusion genes cause cancer
>> in humans (chronic myeloid leukemia in the case of Bcr/Abl and
>> medullary thyroid carcinoma for Ret/PTC). Bcr/Abl is >1500 aa long,
>> Ret/PTC is > than 1200 aa, depending on isotype of Ret.
>>
>> Wot you think?
>
>Mutations such as the BCR/ABL translocation between chromosomes 22 and
>9 are very interesting. Bcr-Abl is a chimeric protein usually
>consisting of the amino terminal 927 amino acids (p210) or the amino
>terminal 426 amino acids (p190) of Bcr (breakpoint cluster region)
>fused to the second exon of the tyrosine kinase Abl (there are also a
>few other potential though less common breakpoints in the Bcr region
>and even in the Abl region).
[big snip of reasonable summary of BCR-ABL, except for the fact
BCR-ABL is still under inactivation control, and the inhibitor peptide
loop still flips between the active and inactive form. In fact, the
BCR-ABL inhibitor and anti-cancer drug glivec works by binding to the
INactive form of BCR-ABL and freezing it in the inactive form]
>In short, the bcr-abl chimeric protein is not a new type of function
>nor is it a beneficial enhancement of an old function. It in no way
>defeats my claim that novel beneficial protein functions above a few
>thousand amino acids cannot evolve via mindless evolutionary
>processes.
See, I predicted this "it is not complex enough" response didn't I.
But here is Dr. Pitmans original claim
> > I have predicted that no mindless process will ever be
> > able to create anything new within a given level of complexity or
> > beyond in real time. I have drawn this line at several thousand amino
> > acids working at the same time.
His current response does involve a slight shifting of goal posts. In
the original statement it was "Anything new" which is not the same as
"novel beneficial protein functions".
Be that as it may, there is an equivocation over "novel function". Dr.
Pitman feels BRC-ABL doesn't count because it is not a "new type of
function", BCR-ABL is a tyrosine kinase link ABL is. However, it has
new targeting (cytoplasm) and new substrates (which plays a
substantial role in it being a cancer gene). For many of the systems
in organisms, novelty comes not from changes in reaction mechanism,
but from targeting and substrate changes.
As a case in point, the lactase which Dr. Pitman bangs on about has
the same reaction _mechanism_ as sugar metabolizing enzymes, but can
accommodate a different substrate (beta-galactoside sugars).
This raises another issue, Dr. Pitman repeatedly confuses size with
"complexity". Lactase is a big enzyme, but it doesn't do anything
particularly complex. When Dr. Pitman says
>Yes, like the lactase and nylonase enzymes, which requires no more
>than 480 or so amino acids in a fairly flexible order at minimum.
>Nothing much higher than this level of complexity has ever been shown
>to evolve.
We are left confused as to what he means "nothing much higher than
this level of complexity has been shown to evolve", if size is not an
issue, why has he drawn a size limit (and the BCR-ABL gene shows that
1,800 aa long functional enzymes with different targeting and
substrate functions can evolve). What does he mean by "complexity"?
Able to perform multiple reactions? There are many enzymes that have
been evolved that perform multiple reactions (Bary Hall has
experimentally evolved such enzymes). What then?
What about "molecular machines" like the flagella and the F1F0 atpase.
They don't fit his criteria, as they are not large proteins, but
non-covalent associations of multiple small subunits. Even the
impressive, long flagellar whip is not a single, big protein, but made
up of thousands of much smaller subunits which self assemble in a
non-covalent manner into the long structure.
So the whole issue of "several thousand amino acids working at the
same time." is completely irrelevant.
Cheers! Ian
=====================================================
Ian Musgrave Peta O'Donohue,Jack Francis,Michael James and Andrew Thomas Musgrave
reynella@RemoveInsret_werple.mira.net.au http://home.mira.net/~reynella/
Sean Pitman
below):
sweetnes...@yahoo.com wrote in message news:<4d71d185.03120...@posting.google.com>...
> I have presented you with proof (laboratory proof, in many instances)
> that your statements are incorrect.
You have shown that short amino acid sequences can come together to
form a new unified function that is indeed unique and of greater
complexity. The only problem you have here is that the minimum amino
acid number required for your most complex example is only 3 or 4
hundred amino acids. My argument is that evolution becomes more and
more difficult the greater the minimum amino acid requirement until it
becomes impossible this side of zillions of years when the minimum
requirement reaches a few thousand amino acids in fairly specified
order. Where are your examples of evolution requiring such a level of
minimum amino acid specificity? I see a lot of hot air coming from
you, but no such example. Where is the reference for such a
demonstration? And, by the way, your lame bacterial swarming example
doesn't even come close (see discussion below).
Recently I have been referred to the example of the bcr-abl chimeric
protein that causes various forms of leukemia such as chronic
myelogenous leukemia. The reason for this referral is the fact that
the bcr-abl function is generally realized with the use of between
1000 and 1800 amino acids. This is good thinking since it is actually
an attempt to controvert my hypothesis whereas you haven't even tried
to present the de novo evolution of anything with a minimum function
requiring over a few hundred amino acids working together at the same
time.
But, before you become excited about the bcr-abl function, there is a
little problem in that it really isn't a new type of function or is it
a selectably beneficial function, but is based on the deregulation of
the pre-established kinase function of the abl protein. For a more
detailed discussion of bcr-abl function see:
http://groups.google.com/groups?hl=en&lr=&ie=UTF-8&selm=80d0c26f.0312060737.24b46c30%40posting.google.com
> I have provided you with at least three examples of exactly that
> happening: simpler systems combining to produce a novel, complex
> system with a different, novel function. The fact that there are more
> ways to combine systems nonfunctionally does not have a bearing on
> this.
Actually it has everything to do with this. What you have shown are
simpler systems combining to produce a novel function that requires,
at minimum, only 3 or 4 hundred amino acids. You have also provided
examples of a simple function requiring less than a few hundred amino
acids at minimum evolving a new type of function within the same level
of complexity, but not anything much greater. Please, I think you can
at least try and do better than this.
> The system does not work by searching randomly through
> thousands of different combinations of amino-acids, until it "hits"
> the right combination. This has been explained to you over and over
> again. It simply does not happen, and your refutation of this does not
> refute anything that is a part of evolutionary theory.
You haven't explained how my ideas are wrong in this regard. All you
have shown is how the same type of function can be up-regulated and
down-regulated (See discussion below of Myxococcus xanthus swarming
evolution demonstrated by Gregory J. Velicer and Yuen-tsu N. Yu of the
Max Planck Institute for Developmental Biology) but not how new types
of function within higher levels of complexity can be evolved without
searching randomly through a very large non-beneficial sequence space.
Sequences with the same types of functions can be clustered around
each other like little islands of stepping stones - very much like a
sentence who's individual letters can be rearranged somewhat without a
complete loss of beneficial meaning, but who's type of meaning is
fairly isolated from other sentences with different types of meaning
or significantly different ways of achieving the same meaning. How to
cross the non-beneficial gap between these two different types of
meaning? That is the question? For shorter sequences, the random
walk required is easier to overcome. However, with each amino acid
increase in the minimum random walk required to achieve the success of
any new type of beneficial function within that level of functional
complexity, the time required grows exponentially.
> Please show me how would it be possible for a change in biological
> systems, such as it happens NOW, at this moment, to go on for, say,
> ten thousand years, WITHOUT producing a novel, complex system?
Ok - take, for example, a particular function that requires, at
minimum 5,000aa at minimum to be realized. Say this sequence happens
to get duplicated so that it can undergo various mutations without
risking significant loss to the original beneficial function (which
has been optimized for its host by now - as far as *level* of function
is concerned). The sequence space at this level of complexity is more
than 10e6500 sequences. The question is, out of all of these
universes of possibilities, how many are or would be beneficial to the
given organism in question? If the organism could use a million
different types of functions to some benefit at this level of
complexity and if each of these million different types of functions
had at least 10e1000 different sequences with at least some selectably
advantageous level of function, then there would be 10e1006 total
beneficial sequences in sequence space with a different type of
function. The question is, how long, on average, would it take to
find any one of these other beneficial sequence islands? Well, you
have to think of ratios at this point. What is the ratio of
beneficial vs. non-beneficial sequences in sequence space? Well,
10e1006 divided by 10e6500 equals 10e5494. That means that for every
one beneficial sequence there are literally universes of
non-beneficial sequences (i.e., there are only around 10e80 atoms in
the entire known universe). With a mutation rate of one mutation per
sequences of 5000aa per generation in a population the size of all the
bacteria on earth, it would literally take zillions of years on
average for even one member of the population to find even one
beneficial sequence with a new type of beneficial function at this
level of complexity.
Of course, you will say that evolution doesn't work like this.
Evolution works by getting various fully formed and functional
sequences to get pasted together to gain a new collective function of
higher complexity. What many don't realize is that the same
statistical problems are involved here. Finding a way to bring a
bunch of small words or even big words totaling 5000 letters together
where each step is beneficially selectable, is much more difficult
than most people realize. The same thing is true for functional base
pair or amino acid sequences. Just because the right sequences are
there as parts of various larger systems of function does not mean
that they will be clipped out and pasted together in just the right
way to make a new system of function that is actually selectably
beneficial at such a high level of functional complexity.
You certainly haven't been able to provide me yet with any such real
time demonstration. I'm still waiting . . .
> I am only now begining to understand the depth
> of what you are saying. The depth of ignorance in it, that is. I
> didn't understand before that you believe that multiudomain proteins
> are just "as complex as the most complex domain" *literally*. Gods.
> And, to top this, you don't belive in the (slightly more intelligent)
> idea that funcitonalities have to evolve suddenly, in a poof, from
> proteins searching through all possible random combinations; you
> actually believe that long proteins with functions can never arise?!?
A bit of clarification: I am well aware that many multi-domain
proteins use all or many of their domains at the same time for a
collective function. However, when you are taking about a particular
function, multiple domains may not be required to achieve this type of
function. The question is, what is the minimum amino acid part
requirement to achieve a particular type of function to a level where
it becomes selectably advantageous? I'm not asking how one can make
the most complex Rube Goldberg-type mousetrap here. I'm asking how
one can make the most simple, bare bones, function of a particular
type that will work to at least some minimal degree of selectable
advantage. Once you have this function, refining it by increasing or
decreasing its level of function (i.e., more or less lactase ability)
is not a problem.
Now, if you can add to this minimum part requirement to achieve a new
type of function with a higher minimum part requirement (i.e.,
bacterial motility), then you will really have something. So far, you
have yet to do this.
> There are hundreds of observed instances. Hell, just last week we had
> a seminar on two insect proteins that mutated (independently), and
> formed a complex with third one; the total complex is over five
> thousand residues long. It performs a function totally unrelated to
> any of its three constituents (two constituents are structural
> proteins, one is, of all things, a kinase; new function has to do with
> mitochondrial RNA repair).
Interesting. Give me the details of this demonstration and/or the
reference for the publication of this demonstration. Does it involve
the deregulation of the kinase function, as is the case with the
chimeric bcr-able function mentioned above?
> >If such differences really do exist between such
> >otherwise similar creatures, then yes, such systems
> >could not evolve in trillions of years and so they
> >must have been designed. I cannot comment specifically
> >on the differences between chickens and turkeys, but I
> >can on the differences between several other life forms.
>
> Such as men and monkeys, eh? Of course.
As I have said over and over again in this forum, I do not think we
know enough about the functional genetic differences between humans
and apes to rule out the possibility of common origin with the use of
genetics alone.
> By your standards, practically every species on earth was designed
> separately and recently. This makes you a YEC. How do you explain away
> the fossil record, the geological data, the radiological evidence,
> cosmological/astronomical evidence, etc?
The definition of "species" is rather subjective. I do not believe
that all of what are now referred to as "separate species" where
designed separately - although I do believe that they were all here
relatively recently. The fossil record and geologic data, to include
the radiological evidence, is not nearly as solid in support of the
long ages of deposition that evolutionists claim it is. It shows very
clear evidence of rapid deposition over the course hundreds and
thousands of years, but certainly not anything even close to a
million, much less a billion years.
See: www.naturalselection.0catch.com
> >These sequences do not have beneficial functions without
> >the cell being what it is - in other words, it defines
> >such sequences as beneficial based on many other systems
> >of function that recognize and use such short beneficial
> >sequences in a beneficial way.
>
> Incorrect. Single amino acids and short polypeptides can (and do)
> catalyze various reactions that could be useful to the proto-organism.
When such sequences are beneficial (which not all of the possibilities
are), they are useful only because the system of the organism
"recognizes" them as useful in a particular environment. Not all such
sequences, even short sequences, are useful to all organisms. Just
because a particular sequence might be useful in one particular
organism does not mean that all organisms will also recognize it as
beneficial.
> >Try evolving this short word, one letter at a time, into
> >a longer and longer word or phrase. See how far you can go.
>
> Quite long, actually. Since words get copied, not just single letters,
> I can get various phrases in.
Try it. Give a demonstration here. Provide us with an environment or
situation. Start with a short word that would be beneficial in such a
situation. Then, adding letters to that word, evolve it were each
addition makes beneficial sense in that situation. Spelling does not
have to be exactly correct, just understandable in a beneficial way.
See how far you can go. Try it here in this forum and include your
demonstration in your reply to this post.
Plus, proteins are not words.
No, they are not. They are even a lot more flexible in their
functional sequencing. But they are very similar in that they are not
completely flexible in their functional sequencing. There is a
minimum limit to the size and order required by a protein-based
function - beyond which all beneficial function suddenly ceases. This
characteristic creates the neutral gap problem just like it does for
English words, sentences, and paragraphs.
> >Again, a multi-domain protein is no more complex in
> >functional ability than its most complex functional domain.
> >The functional domains work independently. It is like multiple
> >proteins stuck together. But, these domains do not necessarily
> >work at the same time to give a greater unified function.
>
> This is perhaps the best example of how poorly you understand
> biochemistry. Considering that your pet lactase is a multidomain
> protein, I wonder how I didn't catch on to this earlier (I know why,
> actually: I gave you far too much credit, and couldn't believe that
> you would claim what you claim without knowing at least some basic
> things).
You will notice that I use the word, "necessarily". A many types of
protein functions do not necessarily require multiple domains in order
to be realized. You yourself admitted as much in your first post when
you claimed that the lactase function is actually much simpler than
its usual 1000 acids would make it appear.
> Practically every large protein is formed of multiple domains, which
> ALL work TOGETHER to produce the function of the protein.
Certainly this is true. If you can present the evolution of such a
collective function that requires, at minimum, all of the various
amino acids in all of the various domains of a particular protein
working at the same time, then you will have something. You have
tried to do this, but your total number of specified amino acids falls
rather short - being less than 500aa so far. You claim to know of
functions that require much more than this, but you have yet to
provide me with the details of such observations and/or a reference to
support such claims.
> One domain
> recognizes the substrate, another may carry the active site of
> catalysis, the third may be involved in carrying the product (or
> recognizing and binding a coenzyme or second substrate...). If you
> don't have specifically defined domains, you have so-called motifs,
> which are smaller sub-domain like structures.
Yes - I agree and know full well about such collective functions. I
think you have misinterpreted what I was trying to say.
> Large proteins that do not work like this are *extremely* rare (which
> is to be expected).
Certainly . . .
> >Beyond a few hundred amino acids required at minimum, such
> >functions simply do not evolve.
>
> I gave you examples of them evolving.
Where? You made a blanket statement but you provided no details or
references to support this statement.
> Sean. I am a biochemist. I work with this stuff every day. How stupid
> do you think I am? You are trying to teach me how to do my job, and
> you are not very good at it.
I don't think you are stupid and I do not assume to teach you how to
do your job. However, I highly doubt that your job invokes the use of
evolution beyond the lowest levels of functional complexity. You
claim that it does or that you do in fact observe such levels of
evolution all the time, but you have yet to support such claims beyond
those that require more than a few hundred amino acids. What I am
talking about here are those types of beneficial functions that
require, at minimum, several thousand amino acids working together at
the same time in a rather specified order.
> Novel proteins of sizes measured in thousands of kD have arisen
> through evolution, in laboratory conditions.
References?
> Well, since the examples I gave you so far simply don't
> count for some reason (mostly that you refuse to accept that things
> happen if your theory says they cannot happen), let me give you a
> relatively recent one. Velicer and Yu, at Max Planck institute in
> Germany; the bacteria, under pressure, developed a complex
> etracellular fibril matrix, and adapted it to achieve a form of
> primitive cooperation.
Actually the M. xanthus bacteria did not develop their extracellular
matrix de novo. They were in fact already able to make this matrix
via the use of pre-existent genes. What happened is that when Velicer
and Yu deleated the genes that gave these bacteria that ability to
achieve S-type swarming, they increased their ability to make more
matrix necessary for A-type swarming. What they evolved was the
ability to make "enhanced quantities" of a matrix which they already
made. In fact, genetic mutants of the evolved strains were constructed
by Velicer and Yu that were unable to make this fibrilar matrix even a
little bit, and guess what, these bacteria never could evolve any type
of significant swarming ability, much less the A-type or S-types.
"Both the genetic and chemical inhibition of fibril-matrix
construction inhibited the evolved swarming phenotypes."
http://www.eurekalert.org/pub_releases/2003-09/m-wsb090503.php
Basically, this is no different than the evolution of penicillin
resistance in bacteria who already have the gene for penicillinase and
all they need to do is make more penicillinase enzyme by decreasing
suppression of penicillinase production. The evolution of decreased
suppression of production of a fairly complex enzyme like
penicillinase is not nearly as difficult as is the evolution of the
penicillinase code to begin with. A single point mutation (many
different ones) can decrease suppression of production. The same
thing happened with this case of M. xanthus swarming evolution.
Please! You really should try to do better than this! I really
thought that you might have it more together - being a "biochemist"
and all. Come on now, is this the very best that you have to offer?
Perhaps you don't like to list the actual references to research
studies like this because you know how very limited they are and that
they really do not support your contentions and most exaggerated
statements in the least.
In any case, this is all the nonsense I have time for today. Hope you
come up with something much better by Monday - as promised. Good luck
. . .
> M.
Ian Musgrave & Peta O'Donohue
Address altered to avoid spam, delete RemoveInsert
On Sun, 7 Dec 2003 18:11:52 +0000 (UTC),
seanpi...@naturalselection.0catch.com (Sean Pitman) wrote:
>This is a repaste of a reply to very similar sweetnes comments (linked
>below):
>
>sweetnes...@yahoo.com wrote in message news:<4d71d185.03120...@posting.google.com>...
>> I have presented you with proof (laboratory proof, in many instances)
>> that your statements are incorrect.
>
>You have shown that short amino acid sequences can come together to
>form a new unified function that is indeed unique and of greater
>complexity. The only problem you have here is that the minimum amino
>acid number required for your most complex example is only 3 or 4
>hundred amino acids.
Coincidentally the median protein size is around 200-250 aa's, so the
amjority of proteins are covered in those kinds of experiments.
>My argument is that evolution becomes more and
>more difficult the greater the minimum amino acid requirement until it
>becomes impossible this side of zillions of years when the minimum
>requirement reaches a few thousand amino acids in fairly specified
>order.
Your argument is wrong, and your "minimum requirment" is irrelevant to
evolutinary biology.
>Where are your examples of evolution requiring such a level of
>minimum amino acid specificity?
Evolution doesn't "require" any such thing. Most protein function
revolves around between 6-12 amino acids, the rest of the protein is
just origami to get those amino acids close to each other. An example
is the serine proteases, where you can use almost any sort of scaffold
to get ser, his and agr close to enough each other for function.
[snip]
>Ok - take, for example, a particular function that requires, at
>minimum 5,000aa at minimum to be realized.
Name one (aside from collagen that is <evil grin>)
[snip]
Deaddog
"Sean Pitman" <seanpi...@naturalselection.0catch.com> wrote in message
news:80d0c26f.03120...@posting.google.com...:
> simpler systems combining to produce a novel function that requires,
> at minimum, only 3 or 4 hundred amino acids.
You do realize, of course, that upon writing these words you lose all
credibility? The difference between experiments done with 300-400 amino
acids that apparently meet your ever-changing, ever-vague criteria for
'complexity' and experiments done with >1,000 amino acids is, well, not very
important. Now, others have tried to clue you in to this, without success.
That's fine. What you believe or don't believe really has little relevance
to scientific progress. But as this inane exchange continues, I hope that
everyone remembers what Sean wrote, above: that he accepts that novel
function can come about by evolution of 300-400 amino acids. Now, let's see
whether we can up that limit a bit:
As the ever-moving pen of science continues to write, any perceived 'gap'
will be closed to quickly as to make you blink.
Blink.
Why, yes, Sean, I did post about this very topic a few days ago. I'm
surprised you missed it. Since you are fond of posting responses consisting
of "Maybe I already said something, let me refer you to my other, garbled
post; did I say something?" let me return the favor, but let me actually
just post the arguments so that it is obvious they meet your criteria.
For the lovely little stageshow that is about to follow, realize that the
approximate sizes of the players are:
LacI = 363 amino acids
TetR = 219 amino acids
cI (lambda repressor) = 237 amino acids
GFP (green fluorescent protein) = 239 amino acids.
That's right, the sum is greater than 1,000 amino acids, and at the
conclusion of this stirring tale a novel function will have been generated
by evolution.
Originally under the title: Synthetic Biology and Intelligent Design:
"It is a slow day pushing the mop here; students are engaged in studying for
finals, and heap their trash politely in the corners.
So it seemed time to address some of Pittman's and others' more ridiculous
claims about the inability of complex systems to evolve de novo. The ironic
thing is not that such systems cannot evolve, but that once the basics of a
regulatory network appear in metabolism the evolution of complexity likely
cannot be stopped.
These findings come under the general rubric of the fields now known as
'systems biology' or 'synthetic biology' (wonderful buzz words, which is why
I am now the co-director of the Center for Systems and Synthetic Biology;
here at Texas we leave no buzzword unused; applications by those with big
brains and / or bags of cash encouraged). Now, while it is clear that
Intelligent Design has so far attempted to squat, Sumo-like, in that
territory previously relegated to the God of the Gaps (as others more
eloquent than myself have stated), the rapid advance of these fields is
making many of the gaps much smaller and more uncomfortable for the
squatters.
A relatively useful case in point is the most excellent paper by Guet et al.
(2002). "Combinatorial Synthesis of Genetic Networks," Science 296:1466.
This paper follows up on earlier efforts in which researchers by and large
attempted to 'intelligently design' new regulatory circuits that could
oscillate, serve as chemical band-pass filters, communicate with one
another, or otherwise exhibit a programmed phenotype. In so doing, it can
be easily argued that the machinery of genetic networks is (a) modular and
(b) can therefore be readily adapted by classic engineering methods.
Indeed, this is in part the purpose of synthetic biology: the put biology
on the same footing as, say, CMOS design (obligatory "heh, heh, heh" to all
old-school organismal biologists; banging shoe on table, we will bury you).
However, what do such design efforts say about the ability of complex
regulatory networks to evolve in the first place? Well, the ease with which
phenotypes could be coaxed from cells that contained rationally engineered
modular parts strongly argued that random associations of the modular parts
should also readily lead to complex regulatory networks. Which is what Guet
et al. (2002) demonstrated.
These researchers took several different promoters / operators and the
repressors (lacI, lambda cI, tetR) that bound to these operators, and
randomly assorted them in different combinations. This is truly the sort of
'experiment' that one can envision occurring during the course of evolution,
as regulatory proteins and their binding sites moved about in genomes either
through sequence evolution (of the rather short operators) or recombination.
What behaviors were observed once the modular genetic regulatory machinery
was randomized and screened? Was it just a hodgepodge of useless signaling,
essentially the equivalent of static or whitenoise? Not at all.
"Altogether, 5^3 = 125 different networks are possible .... [A] fourth
transcriptional unit [was also added], in which the green fluorescent
protein (GFP) expression was controlled by the lambda cI repressible
promoter. The fluorescent signal acts as the network 'output,' whereas the
levels of the two chemical inducers [lactose, for lacI and tetarcycline, for
tetR] were used as 'inputs' .... [We] searched the library for circuits in
which the output is a binary logical function of both inducers. Examples of
such 'logical circuits' are NAND, NOR, or NOT IF [Many, many examples shown
in Figures 2 and 3; based on the characterizations shown, some 30 of the 125
original possibilities demonstrated coherent, complex responsivities.]"
Random mutations also figured into the evolution of some of these complex
phenotypes: "We found a low level of point mutations, which, in some cases,
modify the logical behavior of the networks."
For the most part, though, it was the gears and levers working together
flawlessly following positional randomization that led to the interesting,
emergent, and complex phenotypes. "... connectivity between different
genetic elements varies from network to network so that 13 different
'topologies' can be distinguished .... [S]ingle step changes to the network
connections, in which one promoter replaces another, frequently converted
network operation from one logical function to another."
And then the beautiful, beautiful clincher. They say it much better than I
could:
"From an evolutionary point of view, this observation suggests that ONCE A
SIMPLE SET OF GENES AND CIS-REGULATORY ELEMENTS IS IN PLACE, IT SHOULD BE
POSSIBLE TO JUMP FROM ONE FUNCTIONAL PHENOTYPE TO ANOTHER USING THE SAME
'TOOLKIT' OF GENES by modifying the regulatory connections. Such
discontinuous changes, different from the more gradual effects driven by
successive point mutations, may be achieved in evolution by natural
combinatorial mechanisms like transposition, recombination, or gene
duplication." (emphasis obviously mine; Science editors frown on that sort
of thing)
Wow. Breathtaking. The modularity of the transcriptional, signal
transduction (other papers), and other regulatory machineries made the
invention of complex phenotypes unstoppable.
This is why it's sort of fun to watch the Behe's of the world. It's rather
like looking back at the "If man was meant to fly he'd have wings"
controversy (actually, John, can you turn me on to a spiffy source of quotes
/ analyses relating to this early Waterloo of the Luddites? It would be
helpful in other endeavors).
First there was the 747 / tornado nonsense ... then it was shown that
function could readily emerge from random sequence, given fitness and
replication (duh). So, Behe just retreated to a different Gap, saying that
it wasn't fair (whine) because proteins enzymes did all the work (his rather
sad 'hedgehog on a highway' analogy). So, it was shown that randomization
of nucleic acid enzymes themselves could lead to emergent function, and
Behe's gap got smaller still. We'll have to wait until "Darwin's Cabinet of
Horrors" is released before we can get the next pithy analogy ("molecular
evolution ... is like a monkey paw with three wishes!").
Then there was the supposedly time-tested challenge that multi-protein
systems could not evolve, because of the complexity of the interactions that
would be involved. Except that, as we have seen, the proteins that tend to
be favored in evolution over time are proteins (or circuits) that can
inherently work well together (on another note: evolution of evolvability;
also, think about SH2 / SH3 domains; two-component systems; G
protein-coupled receptors and their attendant G proteins). Modularity of
the regulatory machinery makes it sooooooo much easier to evolve new
function.
Oh, don't worry, they'll just retreat to a new Gap. They always do. I'm
sure Pittman will follow-up with some inane Gapology (whine, but where did
the original networks come from? Answer: they evolved, based on simple
nucleic acid:protein interactions, just as an evolved beta-galactosidase
(ebg) regulon was shown by Barry Hall to evolve from cryptic genetic
material once the regular machinery had been deleted).
But the writing is on the wall, folks. Intelligent Design has arrived on
the scene just in time to be nuked by synthetic biology, which in a
different world might have been called ... intelligent design (unassuming
small letters), except for the fact that evolutionary engineering is still
one of the most powerful tools available for generating phenotype.
And you just wait until we get going with you. Worms that live ten times
longer? Fuck that! I want my, I want my, I want my siRNAs!
Robble, robble, robble.
Non-woof"
Now, Sean, while we all expect that your will sputter incoherently about
what you think is or is not complex enough, and what is or is not evolution,
even you should see that the writing is very much on the wall. You have
already been presented with numerous examples where proteins with novel
functions can evolve like *that* (snap!). You have even come to accept such
examples (see your admission of defeat, at top). Thus, your further attempt
to plant God in a gap that is now, by your own admission, the difference
between 400 amino acids and 1,000 amino acids is really doomed to failure.
And, as I and others have pointed out: you've already failed.
Non-woof
RobinGoodfellow
> sweetnes...@yahoo.com wrote in message news:<4d71d185.03120...@posting.google.com>...
[snip]
>>Well, since the examples I gave you so far simply don't
>>count for some reason (mostly that you refuse to accept that things
>>happen if your theory says they cannot happen), let me give you a
>>relatively recent one. Velicer and Yu, at Max Planck institute in
>>Germany; the bacteria, under pressure, developed a complex
>>etracellular fibril matrix, and adapted it to achieve a form of
>>primitive cooperation.
>
> Actually the M. xanthus bacteria did not develop their extracellular
> matrix de novo. They were in fact already able to make this matrix
> via the use of pre-existent genes. What happened is that when Velicer
> and Yu deleated the genes that gave these bacteria that ability to
> achieve S-type swarming, they increased their ability to make more
> matrix necessary for A-type swarming. What they evolved was the
> ability to make "enhanced quantities" of a matrix which they already
> made. In fact, genetic mutants of the evolved strains were constructed
> by Velicer and Yu that were unable to make this fibrilar matrix even a
> little bit, and guess what, these bacteria never could evolve any type
> of significant swarming ability, much less the A-type or S-types.
> "Both the genetic and chemical inhibition of fibril-matrix
> construction inhibited the evolved swarming phenotypes."
>
> http://www.eurekalert.org/pub_releases/2003-09/m-wsb090503.php
*Screeeech*... WARNING: goal post shift in progress! Please stand
clear of logical contortions!
Sean, it seems to me that this is exactly an example of a function that
you claimed evolution would not be able to produce "this side of a
zillion years". It is clearly a new type of function - *Swarm Motility*
- that the bacterial colony has not posessed previously. It is
unquestionably highly complex, requiring interactions of many thousands
of amino acids in the same time - far more complex, in terms of such
interactions, than, say, the oft-touted bacterial flagellum. It is even
irreducible: knock out either the gene coding for A motility, or the
gene coding for the fibril matrix, and *Swarm Motility* is broken. And
yet, somehow, it evolved.
The reason I'm emphasizing *Swarm Motility* is because it is now a
separate function, much like *Individual Motility* (as carried out by
the "impossible-to-evolve" flaggellum) would be a brand new function by
your definition, regardless of whether the proteins required to attain
this function were involved in other activities beneficial to the cell.
Now you claim that the components for this function were already
there, so it would not be too difficult for evolution to produce the
necessary modification to attain *Swarm Motility*. However, in other
threads, you've claimed that merely having the components is not enough
- that evolution alone could never arrange them in the right order to
achieve the requirement of "a minimum number of amino acids working
together at the same time in the right orientation", such as required
for, say, *Individual Motility*. So, which one is it, then?
Again, to re-iterate, here we have a highly complex, irreducible, novel,
beneficial function which was somehow perfectly content to evolve
without tinkering by any Intelligent Designers. It was arrived at via
straightforward modifications to existing functional gene products.
This is exactly how the evolution of any function, complex or simple, is
supposed to occur. There were no vast neutral gaps to cross, no
functional components to poof out of thin air, no guidance by a designer
to make sure that the pre-existing parts of the system would mesh
together in just the right way to attain the new function. Look at it
any way you like, but your challenge has been met.
(If you're interested in more details about this experiment than can be
found in the press-release, see Nature, v. 425, pp. 75-78, Sep. 2003.)
Finally, your insistence that the extra-cellular matrix evolve "de novo"
for this to be a valid example that meets your challenge underscores a
very serious misunderstanding on your part - one that others on this
newsgroup have pointed out time and again, but which you refuse to see.
In short, you will accept evolution, if, and only if, you see the
molecular equivalent of a dog (or a virus) turning into a cat in a few
years' time. Evolution of systems, as Von Smith in particular pointed
out to you time and time again, can only work on existing materials,
altering it in this way or that, once in a while stumbling onto novel
beneficial effects resulting from modification of existing components.
*Swarm Motility* is exactly such an effect of modifications to the
system consisting of the extra-cellular matrix and the individualistic
A-type motility. Likewise, the less complex (by your metric)
*Individualistic Motility* is likely an effect of modifications of a
pre-cursor secretory system, although in this case we still lack the
exact knowledge of pre-cursor systems and environmental conditions
necessary to describe such modifications with certainty. One thing *is*
pretty certain, though: no "de novo" poofing was required. If such
poofing were to be observed in nature on a regular basis, then maybe you
would find evolution convincing: evolutionary biologists, however, would
be ditching the theory in droves, or modifying it beyond recognition.
Cheers,
RobinGoodfellow.
Ian Musgrave & Peta O'Donohue
Address altered to avoid spam, delete RemoveInsert
On Sun, 7 Dec 2003 18:11:52 +0000 (UTC),
seanpi...@naturalselection.0catch.com (Sean Pitman) wrote:
>sweetnes...@yahoo.com wrote in message news:<4d71d185.03120...@posting.google.com>...
[snip]
>> Please show me how would it be possible for a change in biological
>> systems, such as it happens NOW, at this moment, to go on for, say,
>> ten thousand years, WITHOUT producing a novel, complex system?
>
>Ok - take, for example, a particular function that requires, at
>minimum 5,000aa at minimum to be realized.
[snip]
The Sirens of Titin or
Getting heavy with dr. Pitmans hypothesis
Lets have a look at how well Dr. Pitmans ideas are supported by two
big proteins, trabeculin-alpha (5938 aa) and the aptly named titin
(27118 aa). These proteins help refute what remains of Dr. Pitmans
thesis.
Firstly, some background, as more and more novel functional proteins
have been shown to be produced by mutation and selection, and as the
known density of functional proteins in protein space have dealt a
savage blow to Dr. Pitmans concept of neutral gaps, he has since
re-defined his "neutral gap" theory to apply to proteins of several
thousand amino acids long. Given the example of the BCR-ABL gene
fusion product, where a functional 1,800 aa enzyme with a unique
substrate specificity (a tyrosine kinase) was generated by the fusion
of two smaller enzymes, and several other large, functional fusion
proteins, these would seem to falsify this version of the neutral gap
theory as well. Dr. Pitman in response has now stated that "mindless
mechanisms" cannot produce novel, complex functions in proteins
several thousand aa's long.
To evaluate this, lets look at the nature of proteins that are 5,000
aa's long (using the 5000 aa length form his statement above). Are
the novel, functionally complex proteins which show the thumb-print of
the Intelligent Designer hiding in the >5,000 aa peptides?
The first thing to note is that peptides of 5000 aa and greater
comprise around 0.2% of all currently known peptides, so they are
pretty rare. The second thing to note is that they do not comprise
peptides of great functional complexity; they seem to fall into three
broad categories
1) Adhesins, mucins and other cytokeletal elements (basically glue)
2) Polyproteins, viral gene fusion products that are subsequently
chopped up into smaller elements
3) Polyketide synthases
Polyketide synthases are interesting, because functional polyketide
synthases can be anywhere from ~141 to 9,000 aa's long, it is also a
fairly simple enzymic activity, nothing like the complexity of either
the F1F0 ATPase or the flagella (both of which are associations of
small proteins, rather than being one big protein). I'm not sure Dr.
Pitman would want to claim Polyketide synthase as the fingerprint of
the designer, as it is a) simple and b) its sequence can vary up to
around 100 fold, hardly a highly constrained enzyme.
Down at the lower end, you also get RTX toxins (which are also present
below the 5,000 aa boundary), dynein (a cytoskeletal structural
protein/enzyme whose sequence can also vary considerably and seems
based on ATPases) and the rayonidine receptor (basically a ligand
gated ion channel with its subunits fused, very similar to the IP3
receptor and a range of other ligand gated ion channels). There seems
to be no thumbprint of an Intelligent Designer here.
So lets have a look at titin and trabeculin-alpha.
The heart muscle cell contractile system contains the thin and thick
filaments, and a third filament system composed of titin. Titin is the
largest protein known, with a single polypeptide containing 27118
amino acid residues. Titin constitutes ~10% of the contractile
filament protein mass of the heart muscle, making it the third-most
abundant muscle protein (after myosin and actin).
Now, titin has a reasonably simple structure, it is basically made of
three domains, a PEVK domain, an immunoglobulin domain and an N2B
domain. These are found in many different proteins. The major thing
here is that the immunoglobulin domain is repeated many, many times.
Now, immunoglobulins usually bind antigens as part of the immune
system, but in this case the repeated immunoglobulin domains acts as a
spring, making titin elastic. The structure of titin is exactly what
we would expect if large proteins were constructed by fusion (as seen
with BCR-ABL) and internal duplication of gene segments (a kind of
"molecular stutter" of the gene copying machinary as seen with repeat
duplication in Huntingtin).
Trabeculin-alpha is an actin binding protein, part of the cytoskeletal
structure of the cell, at 5938 aa it's hardly in titin's class, but
still a pretty weighty protein. Basically yrabeculin is a fusion of
Spektrin and Plectin, two known cytokeleton proteins, with 60% of the
protein being multiple repeats of the spektrin element. Again, we see
that a large protein is constructed of a fusion and internal repeats,
just as we would expect if standard copying errors are responsible for
generation of new proteins.
To summarize, Dr. Pitmans contention that large proteins occupy parts
of protein space where mutation and natural selection cannot reach is
falsified by the nature of the known large proteins. These proteins
are almost all fusion/repeat proteins, and look exactly what we would
expect large proteins to look like from known mechanisms of gene
mutation. There is no thumbprint of an Intelligent Designer here.
Cheers! Ian
References
Entrez protein database
Sun,Y., et al., Molecular cloning and characterization of human
trabeculin-alpha, a giant protein defining a new family of
actin-binding proteins J. Biol. Chem. 274 (47), 33522-33530 (1999)
Bang,M.L., et al., The complete gene sequence of titin, expression of
an unusual approximately 700-kDa titin isoform, and its interaction
with obscurin identify a novel Z-line to I-band linking system Circ.
Res. 89 (11), 1065-1072 (2001)
Wu Y, et al., Titin: an endosarcomeric protein that modulates
myocardial stiffness in DCM. J Card Fail. 2002 Dec;8(6 Suppl):S276-86.
Von Smith
> This is a repaste of a reply to very similar sweetnes comments (linked
> below):
>
> sweetnes...@yahoo.com wrote in message news:<4d71d185.03120...@posting.google.com>...
>
<snip>
>
> Of course, you will say that evolution doesn't work like this.
> Evolution works by getting various fully formed and functional
> sequences to get pasted together to gain a new collective function of
> higher complexity. What many don't realize is that the same
> statistical problems are involved here. Finding a way to bring a
> bunch of small words or even big words totaling 5000 letters together
> where each step is beneficially selectable, is much more difficult
> than most people realize. The same thing is true for functional base
> pair or amino acid sequences. Just because the right sequences are
> there as parts of various larger systems of function does not mean
> that they will be clipped out and pasted together in just the right
> way to make a new system of function that is actually selectably
> beneficial at such a high level of functional complexity.
>
Take note of this paragraph, and see what Dr. Pitman writes further
down:
> You certainly haven't been able to provide me yet with any such real
> time demonstration. I'm still waiting . . .
<snip>
>
> > Well, since the examples I gave you so far simply don't
> > count for some reason (mostly that you refuse to accept that things
> > happen if your theory says they cannot happen), let me give you a
> > relatively recent one. Velicer and Yu, at Max Planck institute in
> > Germany; the bacteria, under pressure, developed a complex
> > etracellular fibril matrix, and adapted it to achieve a form of
> > primitive cooperation.
>
> Actually the M. xanthus bacteria did not develop their extracellular
> matrix de novo. They were in fact already able to make this matrix
> via the use of pre-existent genes. What happened is that when Velicer
> and Yu deleated the genes that gave these bacteria that ability to
> achieve S-type swarming, they increased their ability to make more
> matrix necessary for A-type swarming. What they evolved was the
> ability to make "enhanced quantities" of a matrix which they already
> made. In fact, genetic mutants of the evolved strains were constructed
> by Velicer and Yu that were unable to make this fibrilar matrix even a
> little bit, and guess what, these bacteria never could evolve any type
> of significant swarming ability, much less the A-type or S-types.
> "Both the genetic and chemical inhibition of fibril-matrix
> construction inhibited the evolved swarming phenotypes."
>
In other words, this example does not disprove your strawman model of
how complex systems evolve because it evolved by some other means than
your strawman model.
The wild strains could not swarm using A-type mobility. The initial
mutants couldn't swarm *at all*. The evolved mutants could. A novel
swarming mobility is, by any reasonable definition, a new function,
and it fits your description of a multi-protein function, etc. And it
evolved the way I and others have said such things evolve: by gradual
modification of a precursor structure, not by popping into being de
novo.
> http://www.eurekalert.org/pub_releases/2003-09/m-wsb090503.php
>
> Basically, this is no different than the evolution of penicillin
> resistance in bacteria who already have the gene for penicillinase and
> all they need to do is make more penicillinase enzyme by decreasing
> suppression of penicillinase production. The evolution of decreased
> suppression of production of a fairly complex enzyme like
> penicillinase is not nearly as difficult as is the evolution of the
> penicillinase code to begin with. A single point mutation (many
> different ones) can decrease suppression of production. The same
> thing happened with this case of M. xanthus swarming evolution.
In a sense, you are absolutely correct that the novel swarming
mobility is "no different" than the de-suppression of penicillinase:
they are not fundamentally different in the evolutionary processes
required to make them happen. And this sense in which you are correct
is devastating to your entire "neutral gaps" argument. The types of
evolutionary changes that do something as simple as de-suppressing
enzyme production can also bring about novel multi-protein functions
of the sort you describe, to include novel modes of motility (remember
that you claimed earlier that all modes of bacterial motility were
complex in this fashion).
And this brings us back to your paragraph earlier in the post, in
which you basically suggest that sufficiently complex functions cannot
evolve regardless of how many of the components are already in place.
Obviously, this example puts paid to *that* particular claim, as you
implicitly acknowledge in your attempt to dismiss it immediately
above. You want to have your cake and eat it, too:
On one hand, you continue to assert that novel multi-protein functions
can't evolve in a "zillion" years because of all the neutral gaps, and
you ignore attempts to explain that you cannot possibly estimate such
"gaps" absent a description of the system's ancestor. On the other
hand, when presented with an example of precisely what people here
have been trying to tell you about, you want to say that this doesn't
prove you wrong, because evolving the new function from its immediate
precursor didn't involve any, or at least not enough, neutral gaps.
I have told you before that purely desciptive differences between two
given functions do not define the "neutral gaps" between them. The
complexity of a given function in and of itself tells you absolutely
nothing about how easy or difficult it is to evolve. What you have to
do is consider possible precursors, and try to find evolutionary
pathways that connect those precursors to the structure or function in
question. Looked at without access to its provenance, this A-swarming
looks just as complex as the flagellum, and yet,as you yourself argue,
evolving it can happen fairly trivially, given suitable precursors.
Von Smith
Fortuna nimis dat multis, satis nulli.
Sean Pitman
> >You have shown that short amino acid sequences can come together to
> >form a new unified function that is indeed unique and of greater
> >complexity. The only problem you have here is that the minimum amino
> >acid number required for your most complex example is only 3 or 4
> >hundred amino acids.
>
> Coincidentally the median protein size is around 200-250 aa's, so the
> amjority of proteins are covered in those kinds of experiments.
It is good to see you back Ian. Hope your interesting life calmed
down just a bit.
In any case, certainly the average single protein size is around 200
to 300aa, but there are many functions at higher levels of complexity
that require more than this average number of amino acids to achieve
minimum beneficial function. Many functions require multiple proteins
working at the same time, or at least larger proteins with multiple
unique domains working at the same time.
> >My argument is that evolution becomes more and
> >more difficult the greater the minimum amino acid requirement until it
> >becomes impossible this side of zillions of years when the minimum
> >requirement reaches a few thousand amino acids in fairly specified
> >order.
>
> Your argument is wrong, and your "minimum requirment" is irrelevant to
> evolutinary biology.
So you say, but your claim that the total number of beneficial
sequences that have a particular function, such as the lactase
function, is less than 10e100 or so is a problem for you. By your own
statements and research into this question it seems like the minimum
amino acid requirement for beneficial lactase function in a living
thing is greater than 400aa. The sequence space at this level is
around 10e520. This creates a ratio of lactase vs. non-lactase
sequences of around 1 in 10e420 - using your own numbers. This ratio
is a problem since many other types of functions would probably have a
similar ratio at this level of complexity. Even if there where a
billion different types of beneficial functions in this sequence space
and even if each one of these types of functions was coded for by
10e100 amino acid sequences, the total number of beneficial sequences
would still be only 10e109. The ratio of beneficial vs.
non-beneficial would still be well over 1 in 10e400. Now that is a
problem for your position. How do you get from one type of function
to a new type of collective amino acid function across such a gap that
grows much much greater than this at higher and higher levels of
minimum amino acid requirements?
> >Where are your examples of evolution requiring such a level of
> >minimum amino acid specificity?
>
> Evolution doesn't "require" any such thing. Most protein function
> revolves around between 6-12 amino acids, the rest of the protein is
> just origami to get those amino acids close to each other. An example
> is the serine proteases, where you can use almost any sort of scaffold
> to get ser, his and agr close to enough each other for function.
Not every protein function is as flexible as a serine protease. Many
relatively simple single protein functions, although requiring fewer
amino acids at minimum, are far more constricted in their sequence
flexibility. Cytochrome c is just one example. It is true that some
positions are far more flexible than others, but many positions are
more constricted to certain categories of amino acids (acidic, basic,
hydrophilic, hydrophobic, etc). Many other positions are restricted
to just a handful of amino acid options, and a few are highly
restricted to just one or two amino acid options. Also, even though
just about any amino acid can be mutated, one at a time, in a protein
sequence without a complete loss of beneficial function, it is much
harder to change too many proteins at the same time without a complete
loss of beneficial function - just as it is for changing the letters
of a paragraph written in English.
The point is that your statement that only 6 to 12 amino acids are
required to be in a particular order, out of 200 to 300 amino acids,
is a significant understatement of the true limitations that are
required for many single protein functions to be realized. These
limitations of the larger protein sequence create a much lower ratio
in sequence space than would otherwise be assumed based on the
accounting of only 6 to 12 amino acids. For example, if the ratio of
lactase vs. non-lactase enzymes were truly based on the correct
orientation of only 6 amino acids, the maximum ratio of lactase vs.
non-lactase enzymes in sequence space would be as high as 1 in 64
million. With such a high ratio of lactase sequences, an average
bacterial colony would be able to evolve a lactase enzyme in less than
one year. This just doesn't happen in real life. Many types of
bacteria have not been able to evolve the relatively simple lactase
function in 50+ years (over a million generations). Oh no, many
protein functions are far more restricted than you seem to understand.
> >Ok - take, for example, a particular function that requires, at
> >minimum 5,000aa at minimum to be realized.
>
> Name one (aside from collagen that is <evil grin>)
Collagen is based on polymer assembly of short proteins in long
strands. Obviously I'm not talking about the assembly of polymers
here. I'm talking about minimum genetic information here. Not very
much genetic information (minimum size of gene requirement) is needed
for collagen production - relatively speaking. For example, all
bacterial motility systems require far more genetic information (i.e.,
minimum base pair sequence requirement) before the motility function
will be realized. This includes the polymer production of structures
such as the flagellar tube etc. Even though the flagellum makes up a
significant percentage (as far as total proteins and amino acids are
concerned) of the total structure of the flagellar motility system,
the minimum genetic information required to code for this system is
what is important. The total number of different proteins required,
not the total number of the same proteins required, is what is
important for determining informational complexity.
Now, I know that you have tried to suggest that bacterial motility
systems can be produced with just two or three different types of
proteins working together at the same time, but you have failed to
provide an adequate demonstration or reference to support this
statement of yours. The fact is that the minimum part requirement for
types of bacterial motility like the flagellar system require more
than 20 different proteins working together at the same time for a
total of well over 5,000aa at minimum.
> Cheers! Ian
howard hershey
> Sean Pitman wrote:
>
>> sweetnes...@yahoo.com wrote in message
>> news:<4d71d185.03120...@posting.google.com>...
>
[snip]
> Actually the M. xanthus bacteria did not develop their extracellular
> matrix de novo. They were in fact already able to make this matrix
> via the use of pre-existent genes.
[snip]
Well, duh. No real biologist argues that evolution works by
poofing "thousands of amino acid" changes into existence all at once.
In fact evolutionary models all *assume* that each intermediate
selectable step in the evolution of *any* system involves modification
of pre-existent genes by mechanisms no different than the ones described
for the M. xanthus extracellular matrix.
Evolution is always descent with *and by* modification of pre-existent
genes (or duplicates, including chimeric duplicates, thereof). These
events can lead to emergent properties that can be selected for that
were not anticipated by anyone and were not any sort of teleological goal.
No real, non-strawman model of evolution starts with a random sequence
nor with a completely unrelated gene (which is the equivalent of a
random sequence).
No real, non-strawman model of evolution requires mutation of "thousands
(or even hundreds or *even* tens) of amino acids" to reach a new
selectable function.
Nor real, non-strawman model of evolution ever requires that all
intermediates show a *necessary* teleological end point function
(present function is always mistakenly thought to have been the
*necessary* teleological end point by creationists) that *must* be
reached.
Real, non-strawman models of evolution involve only functions that
*have* been or *can* be reached by evolutionary mechanisms and functions
that *haven't* been or *can't* be reached. What *you* need to do is
show that there are some functions that *have* been reached in organisms
which *cannot*, even in principle, be reached by *any* stepwise process
with useful intermediates.
In real, non-strawman evolutionary models there are no "large neutral
gap", ever, that actually *need* to be crossed. There are only "small
neutral gaps" which produce a different or modified and *functional*
system that can then be subsequently further modified. This is because
in real, non-strawman evolutionary models, the precursor which gets
changed (or duplicated and changed) to form a new functional
intermediate state is not a random sequence or some completely unrelated
protein, but rather a protein which is *already* close to the proper
sequence and which *already* has relevant functionality (needing to
either only change a substrate, change a function, change in regulation,
or change an association with other proteins to form the modified
intermediate). None of those types of changes requires crossing "large
neutral gaps".
If you could show that there is no possible pathway for a presently
existing system which cannot, even in principle, evolve via such
functional (but not necessarily the current function) intermediate
stages, you would have a real argument rather than a diatribe against a
strawman.
Yet you keep presenting this strawman idea that, somehow, in a way you
cannot clearly explain, by using *words* like "thousands of amino acids"
and "exponential increase in complexity" which are wielded like the
cross at a vampire, claim that there are large gaps without the
possibility of intermediate steps with *any* new selectable function
(not necessarily the teleologic function -- aka, present function -- you
assign to it). Somehow you think that evolution requires mechanisms
that differ dramatically between what you agree that we can see with
antibiotic resistance, which involves short chains of mutational change
before there is selectable effect, and the types of systems you see as
problematic, which you falsely see as requiring long chains of
"thousands" of mutational events before there is *any possible*
selective effect. You do this by invoking, perhaps subconsciously, the
idea that current function is a teleological necessity and the only
selectable function. And that there is no possible selectable function
for intermediate structures. That is, you are making the "What good is
half a wing?" argument in molecular terms, asserting or assuming that
all possible intermediate steps in the formation of, say, bacterial
flagella are utterly useless between some unspecified random sequence
and the current/teleologic function.
It just ain't so. The very reason for looking for possible proteins
from which the proteins involved in bacterial flagella arose (by looking
at sequence similarity) is that these are the likly starting points for
each evolutionary step. Not some random sequence. Not some random
protein in sequence space. Rather you *start with* proteins that are
close to end sequence (and usually function as well) and which do not
require "thousands" of changes to produce the function. That does not
mean that the cells *intentionally* start with these proteins of similar
sequence; that would be teleological. Rather, it is that only these
proteins of similar sequence which *can* evolve to the new or modified
function. If such a *similar* protein did not exist in the ancestral
organism, you simply would not evolve the modified protein.
You are looking at systems where such events *did* happen. That does
not mean that they were teleologically *destined* to happen. You cannot
argue that the observed current result *didn't* happen. What you must
do is show that the observed current result *couldn't* happen, even in
principle. Both of us can agree that it *couldn't* happen naturally if
evolution were to work under the assumptions you propose (you start with
random sequence(s) or random unrelated protein(s); there is no possible
intermediate positions or states of *any possible* functional utility
between that random starting position and the current/teleological
function; there must be thousands of completely random changes before
there can be *any possible* selectable functional utility different from
the original sequence's utility). That is the strawman you have been
tilting against. Now let's see you battle the real thing.
Evolution works for all systems exactly like it does for antibiotic
resistance. Each 'step' in a non-teleological pathway involves
selection for variants (or duplicates with variants) with modified
function via a short (not large) chain of mutational events. The
selectable function at each 'step' need not be the current function.
The only difference in the systems you find problematic (flagella) and
those you don't (antibiotic resistance) is the number of such
independent funtionally useful 'steps' required.
For the case of bacterial flagella, there is evidence that favors its
evolution via modification of a stepwise series of systems originally
designed for a different function -- protein export. But the last major
step in this proposed model, the modification that produced motility of
the 'whip' is just that -- the last major step. There are a number of
other proposed intermediate steps proposed to reach that penultimate
point (penultimate only wrt the current flagellar activity; in some
bacteria, the flagellar activity is probably penultimate to some current
Type III export activity). And there is sequence evidence wrt which
*related* proteins are *modified* to generate these individual steps and
evidence (from actual functions that actual bacteria perform) wrt to
functions that simpler intermediate states could perform.
Your problem is not to merely point at "the flagella" and claim and
handwave that one cannot go from nothing functional to the modern
flagella in one swell foop by generating thousands of specific requried,
but selectively neutral, mutations in random proteins or sequences. We
can both agree that that could not happen (except by supernatural or
unknown mechanism; science would only claim the second alternative).
Your problem is that you need to point out which of the proposed
intermediate states of functional utility *cannot*, even in principle,
be reached from the proposed precursor *related* proteins and the
proposed previous states of functional utility.
To do that, the first step would be to actually read and/or understand
(since you claim to have read it) the articles by Nicholas Matzke and
Ian Musgrave. Their proposals are a tad more detailed (and testable,
that is, having testable consequences) than merely the conversion of a
Type III-like protein to a flagella. And vastly more detailed (and
testable) than a theory that posits "Bacterial flagella gets poofed into
existence at some unspecified time by some unspecified mechanism by some
unspecified agent which is, for some unspecified reason, thought to be
intelligent."
--
Howard V. Hershey
Sean Pitman
> Lets have a look at how well Dr. Pitmans ideas are supported by two
> big proteins, trabeculin-alpha (5938 aa) and the aptly named titin
> (27118 aa). These proteins help refute what remains of Dr. Pitmans
> thesis.
Oh, this should be good!
> Firstly, some background, as more and more novel functional proteins
> have been shown to be produced by mutation and selection, and as the
> known density of functional proteins in protein space have dealt a
> savage blow to Dr. Pitmans concept of neutral gaps, he has since
> re-defined his "neutral gap" theory to apply to proteins of several
> thousand amino acids long.
The neutral gap problem has always been about the density of
beneficial sequences in sequence space. You haven't yet provided me
with any reasonable support for your notion that the density of
beneficial sequences is as high as 10e11 for the 100aa level for
fairly specified proteins (such as cytochrome c and the like) and that
it stays at this density at increasing levels of minimum amino acid
requirements. For example, when I asked you how many lactase
sequences exist in sequence space, you gave a number of around 10e90.
The problem here is that the lactase function seems to require more
than 400aa in a fairly specified order. The ratio that this creates
is truly miniscule at less than 1 in 10e500.
Where again is this "savage blow" that you speak of?
> Given the example of the BCR-ABL gene
> fusion product, where a functional 1,800 aa enzyme with a unique
> substrate specificity (a tyrosine kinase)
This is not a unique specificity! If anything it is a more general
specificity of the abl-sequence.
> was generated by the fusion
> of two smaller enzymes, and several other large, functional fusion
> proteins, these would seem to falsify this version of the neutral gap
> theory as well.
Not at all. As previously described, the kinase activity of the abl
SH1 domain is freed from suppression by the binding of a relatively
short sequence from the bcr-protein (as few as 60aa and probably a lot
less would work). This results in an increase in activity of the same
function that was always there. It also results in a non-beneficial
disease that sickens and kills its host. So, no new function and
certainly no beneficial function results here. Genetic information
was lost here, not gained. A loss of inhibitory control over the SH1
kinase domain on the abl protein is all that happened here. Such a
loss of pre-established functions or interactions are easy to achieve
via many different sequences since it is far easier to destroy than to
create an independent function.
> Dr. Pitman in response has now stated that "mindless
> mechanisms" cannot produce novel, complex functions in proteins
> several thousand aa's long.
You fail to complete my requirements here. Also included is that
these amino acids must be in a fairly specified order. This
effectively rules out those huge proteins that are basically made up
of large sections of repeated sequences. Also, the level of
complexity is determined by the *minimum* amino acid requirement
needed to give rise to a particular type of function - not the maximum
amino acid system which can be found to have a particular function.
For example, E.coli bacteria usually use a lactase gene that codes for
an amino acid sequence of around 1000aa. This subunit sequence
combines with 3 other identical subunits to form the lactase enzyme
with 4000aa total. However, this is not, as you have pointed out so
clearly yourself, the minimum amino acid requirement needed to achieve
the lactase function to at least some selectably beneficial level. The
same thing is true of the evolved "ebg"-lactase enzyme who's gene
produces a subunit protein of around 1000aa that forms a hexamer with
5 other identical copies for a total of around 6000aa. This is not an
example of evolution at the level of 6000aa since the gene only codes
for 1000 of them and of these, only 480 or so seem to be absolutely
required in a fairly specified order for minimum lactase function to
be realized.
Now, do you understand better what my criteria are for sequence
complexity?
> To evaluate this, lets look at the nature of proteins that are 5,000
> aa's long (using the 5000 aa length form his statement above). Are
> the novel, functionally complex proteins which show the thumb-print of
> the Intelligent Designer hiding in the >5,000 aa peptides?
>
> The first thing to note is that peptides of 5000 aa and greater
> comprise around 0.2% of all currently known peptides, so they are
> pretty rare. The second thing to note is that they do not comprise
> peptides of great functional complexity; they seem to fall into three
> broad categories
What you evidently fail to realize is that I'm not limiting myself to
single protein sequences at all. Such huge single protein sequences
are most certainly rare, but multiprotein functions at this level of
complexity are not rare, and they are far more specified in their
sequence structure than are those relatively few examples of mammoth
single proteins - most of which require far fewer amino acids at
minimum to have some selectably beneficial function of the same type.
> 1) Adhesins, mucins and other cytokeletal elements (basically glue)
All of these types of functions have a very low degree of specificity
and a relatively low minimum amino acid requirement.
> 2) Polyproteins, viral gene fusion products that are subsequently
> chopped up into smaller elements
These polyproteins do not work until they are chopped up into their
independent elements.
> 3) Polyketide synthases
Again, the minimum amino acid requirement for this type of function is
rather minimum as you yourself observe (requiring as few as 141aa for
beneficial function of this type).
> Polyketide synthases are interesting, because functional polyketide
> synthases can be anywhere from ~141 to 9,000 aa's long, it is also a
> fairly simple enzymic activity, nothing like the complexity of either
> the F1F0 ATPase or the flagella (both of which are associations of
> small proteins, rather than being one big protein).
Exactly. I'm only talking about the minimum requirement for amino
acids to work together at the same time as either part of a single
protein or many separate proteins that work together at the same time
(as in a flagellar motility system). When multiple proteins work
together at the same time in a specified orientation with each other,
their individual amino acids are all working together at the same time
as well. They can therefore be totaled to find the general level of
complexity involved. This is a reflection of the amount of genetic
real estate that was required to code for such systems of function.
> I'm not sure Dr.
> Pitman would want to claim Polyketide synthase as the fingerprint of
> the designer, as it is a) simple and b) its sequence can vary up to
> around 100 fold, hardly a highly constrained enzyme.
It depends upon what you mean by "100-fold" variability. For example,
the cytochrome c sequence is considered to be highly constrained and
yet you quote Yockey as suggesting that this ~100aa sequence can vary
between 10e60 to 10e90 sequences in a sequence space that is around
10e130 sequences in size. How many fold is that?
> So lets have a look at titin and trabeculin-alpha.
>
> The heart muscle cell contractile system contains the thin and thick
> filaments, and a third filament system composed of titin. Titin is the
> largest protein known, with a single polypeptide containing 27118
> amino acid residues. Titin constitutes ~10% of the contractile
> filament protein mass of the heart muscle, making it the third-most
> abundant muscle protein (after myosin and actin).
Certainly true . . .
> Now, titin has a reasonably simple structure, it is basically made of
> three domains, a PEVK domain, an immunoglobulin domain and an N2B
> domain. These are found in many different proteins. The major thing
> here is that the immunoglobulin domain is repeated many, many times.
Exactly. And this lowers the specificity of the amino acids in the
titin protein significantly. It also dramatically lowers the minimum
amino acid requirement to achieve titin function to at least some
minimum level of beneficial selectability.
> Now, immunoglobulins usually bind antigens as part of the immune
> system, but in this case the repeated immunoglobulin domains acts as a
> spring, making titin elastic. The structure of titin is exactly what
> we would expect if large proteins were constructed by fusion (as seen
> with BCR-ABL) and internal duplication of gene segments (a kind of
> "molecular stutter" of the gene copying machinary as seen with repeat
> duplication in Huntingtin).
Yes, if all protein functions were actually this redundant in sequence
requirements - which they aren't.
> Trabeculin-alpha is an actin binding protein, part of the cytoskeletal
> structure of the cell, at 5938 aa it's hardly in titin's class, but
> still a pretty weighty protein. Basically yrabeculin is a fusion of
> Spektrin and Plectin, two known cytokeleton proteins, with 60% of the
> protein being multiple repeats of the spektrin element.
Exactly. Again you strike with the multiple repeats of the same
sequences. The problem is that this lowers the specificity of the
protein as well as the minimum part requirement for beneficial
function of this type.
> Again, we see
> that a large protein is constructed of a fusion and internal repeats,
> just as we would expect if standard copying errors are responsible for
> generation of new proteins.
Yes, but not all protein function are like this - compose of
relatively few novel subunit sequences repeated over and over in
tandem sequence. There are many functions that require a lot more
amino acid specificity than this. Bacterial motility systems are just
one example.
> To summarize, Dr. Pitmans contention that large proteins occupy parts
> of protein space where mutation and natural selection cannot reach is
> falsified by the nature of the known large proteins.
Perhaps the strawman that you have built to represent my position has
been falsified by these examples, but my actual position has not been
falsified at all. Do you really think that I was unaware of these
examples? Come on now, there are many many more examples of large
proteins composed of polymer-like repeats of relatively few types of
amino acid sequences/structures. These are not examples of high
complexity at all and my original challenge took these into
consideration.
> These proteins
> are almost all fusion/repeat proteins, and look exactly what we would
> expect large proteins to look like from known mechanisms of gene
> mutation. There is no thumbprint of an Intelligent Designer here.
I just knew that someone would try to float something like this, but I
didn't think it would be you. I thought more of you in particular.
Oh well, I still think that you have it more together than most who
frequent this forum.
> Cheers! Ian
RobinGoodfellow
accept evolution now. Now where do I go to get my EAC membership?
(Sorry, I understand you meant this in reply to Sean, but I couldn't
help myself. And sorry again for top-posting, but it seemed easier in
this case. Great post, as usual.)
howard hershey <hers...@indiana.edu> wrote in message news:<br2h4e$s61$1...@hood.uits.indiana.edu>...
Deaddog
"Sean Pitman" <seanpi...@naturalselection.0catch.com> and Floppy come
out for an encore in message
news:80d0c26f.03120...@posting.google.com...:
E.coli bacteria usually use a lactase gene that codes for
> an amino acid sequence of around 1000aa. This subunit sequence
> combines with 3 other identical subunits to form the lactase enzyme
> with 4000aa total. However, this is not, as you have pointed out so
> clearly yourself, the minimum amino acid requirement needed to achieve
> the lactase function to at least some selectably beneficial level. The
> same thing is true of the evolved "ebg"-lactase enzyme who's gene
> produces a subunit protein of around 1000aa that forms a hexamer with
> 5 other identical copies for a total of around 6000aa. This is not an
> example of evolution at the level of 6000aa since the gene only codes
> for 1000 of them and of these, only 480 or so seem to be absolutely
> required in a fairly specified order for minimum lactase function to
> be realized.
>
> Now, do you understand better what my criteria are for sequence
> complexity?
Yes, Sean, we understand completely that you have dragged Floppy, your
strawman friend, from the cornfield and continue to attempt to put him on
display here. Collectively a number of examples have now been provided that
show that evolution can readily create new function at the level of tens to
hundreds to thousands of amino acids. You continue to ask an idiotic
strawman question, which apparently can be boiled down to this: can you get
novel function from a thousand random amino acids?
Note that your idiotic strawman question has nothing whatsoever to do with:
(a) evolution in the distant past, or (b) the evolution of organisms. In
the distant past, as has been explained to you again and again, short random
sequences could easily have generated nascent functions, which in turn
duplicated, diverged, and combined to yield functional metabolism. In
organisms, as has been explained to you again and again, the many extant
proteins can easily evolve to generate new functions. Examples of novel
functions that cover virtually any size range you care to mention have been
provided.
Your continuing to ask your idiotic strawman question is indicative
primarily of your own utter inability to understand anything that is said to
you. Because, bucko, the fact is that your idiotic strawman questions do
not, for example, prevent the real-world evolution of novel enzymes with
interesting functions. For example, having actually worked on the evolution
of galactosidases (Matsumura et al. (1999), Nature Biotechnology 7:696-701;
Matsumura and Ellington (2001), J Mol Biol 305:331-339) it is clear that
many enzymes can readily evolve to use substrates they previously would not
have touched. While you crow about how Hall was not able to evolve
galactosidase function once ebg was deleted, the fact is that the
acquisition of galactosidase function is likely determined by the mutational
space you are willing to examine. The enzyme that we worked on,
beta-glucuronidase, is an E. coli enzyme, and is only a few amino acids away
from functioning as a reasonable beta-galactosidase. The fact that Hall did
not pick up this particular set of mutations is more likely due to the rate
of mutation he used, the timescale allowed for the evolution of function,
and the stringent requirement for growth imposed, rather than to any
unbridgable gap in enzyme sequence or function.
Irrespective of whether you understand that your friend Floppy in no way
represents reality, reality has this really nasty habit of just marching on
(and over you).
To reiterate, in hopes that some portion of this will penetrate your
cranium:
(1) De novo evolution of enzyme function at or near origins almost certainly
involved strings much less than 1,000 amino acids in length. The ability to
generate enzyme function from small peptides has been adequately
demonstrated.
(2) Modern evolution does not involve the evolution of function de novo;
rather, in the context of any reasonably complex genome the alteration of
enzyme function and metabolic regulation by duplication, divergence, and
recombination leads to an almost pluripotent phenotype.
(3) To the extent that the vague and meaningless requirement of 'complexity'
that you continue to set forth without adequate qualification or definition
is actually the statement "it is difficult to derive protein function from a
random thousand amino acids" ... well fucking duh! No one has said or ever
will say that this has anything to do with biological evolution, except
those more interested in blather than accomplishment.
Take Floppy back to the field Sean. He's beginning to smell.
Non-woof
ArWeGod
news:80d0c26f.03120...@posting.google.com...
> more difficult the greater the minimum amino acid requirement until it
> becomes impossible this side of zillions of years when the minimum
> requirement reaches a few thousand amino acids in fairly specified
> order.
Don't you understand? The Universe has had that many years. And probability
doesn't REQUIRE a "zillion" years. It might be unlikely, but unlikely things
can also happen RIGHT AWAY. Think about the lottery. It's very unlikely that
YOU will win, and sometimes no one does, but eventually (like in a few
weeks) SOMEONE does. That's all that happened.
Remember that while it may take "zillions" of years on one planet, the
Universe has "zillions" of planets to try ALL AT ONCE. The time frame then
changes to NOW.
-ArWePetriDishes
howard hershey
Deaddog wrote:
> "Sean Pitman" <seanpi...@naturalselection.0catch.com> and Floppy come
> out for an encore in message
> news:80d0c26f.03120...@posting.google.com...:
[snip]
> To reiterate, in hopes that some portion of this will penetrate your
> cranium:
Not likely.
> (1) De novo evolution of enzyme function at or near origins almost certainly
> involved strings much less than 1,000 amino acids in length. The ability to
> generate enzyme function from small peptides has been adequately
> demonstrated.
>
> (2) Modern evolution does not involve the evolution of function de novo;
> rather, in the context of any reasonably complex genome the alteration of
> enzyme function and metabolic regulation by duplication, divergence, and
> recombination leads to an almost pluripotent phenotype.
Just to give some numbers wrt the role of duplications: This is from a
review published in Advances in Genetics, 46, 20002, but available
on-line at
www.zoology.ubc.ca/~otto/Reprints/OttoYong2002.pdf
"Recent genomic analyses have clarified the extent to which gene
duplications (e.g. Lynch and Conery, 2000; The Arabidopsis Genome
Initiative, 2000) and genome duplications (e.g. Lundin, 1993; Wolfe and
Shields, 1997; Postlethwait et al., 1998; Vision et al., 2000) occur.
These studies have found that a remarkably high fraction of genes are
closely related to other genes within the genome. For example, the
fraction of genes that represent recent duplication events (recent
enough to generate recognizable paralogs) is 11.2% in Haemophilus
influenzae, 28.6% in Saccharomyces cerevisiae, 65.0% in Arabidopsis
thaliana, 27.5% in Drosophila melanogaster, and 44.8% in Caenorabditis
elegans (The Arabidopsis Genome Initiative, 2000). These numbers
increase substantially if less stringent criteria are used to identify
paralogs (compare the above results to those of Rubin et al., 2000).
Indeed, it seems likely that a large fraction, if not all, genes within
a genome are ultimately related by descent to a small number of genes
that arose early in our evolutionary history (Maynard Smith, 1998),
although there is some evidence for the de novo evolution of short
regulatory and signal transduction domains (Chervitz et al., 1998).
Understanding how these ancestral genes have given rise, through
duplication followed by diversification, to the vast number and array of
genes present in extant organisms has fascinated evolutionary biologists
for decades (Haldane, 1933; Fisher, 1935; Grant, 1963; Spofford, 1969;
Ohno, 1970; Stebbins, 1980)."
"Rate of Gene Duplication – What is the rate at which new gene
duplications appear? Lynch and Conery (2000) recently addressed this
question by scouring complete genomic sequences for very young gene
duplicates (excluding multigene families and transposable elements).
They found 10 pairs of genes in D. melanogaster, 32 pairs in S.
cerevisiae, and 164 pairs in C. elegans whose level of silent-site
divergence was less than 0.01. Using estimated rates of silent-site
substitution, they inferred that 31 new duplicates arose per genome per
million years in flies, 52 in yeast, and 383 in nematodes. In the
absence of selection, one can use the neutral theory (Kimura, 1983) to
make two important predictions about gene duplications from these
estimates. (1) The rate at which gene duplicates spread throughout a
population (= a substitution) should equal the genome-wide rate of gene
duplication, v, regardless of the population size. (2) The probability
that any two haploid genomes randomly sampled from a diploid species
will differ in terms of the duplicate genes that they carry is 4 Ne v/(4
Ne v +1). (Aside: Ne is a measure of the size of a population in terms
of how much allele frequencies change by chance from generation to
generation, i.e. as a result of random genetic drift. A population with
a very large census size, N, may nevertheless have a low Ne if only a
few individuals ever reproduce. See Crow and Kimura, 1970, for further
details.) Given their estimates of ?, Lynch and Conery (2000) estimate
that roughly half of the genes within the genome are expected to produce
duplicates that spread through the population over a time period of 35 -
350 million years. For scale, mammals and birds first appeared in the
fossil record around 200 - 250 MYA, which is also the time period during
which monocots and dicots diverged."
It should be noted that this would be an *underestimate* of the actual
rate of gene duplication because it assumes that all duplications are
neutral or beneficial and none are detrimental. If there are
detrimental duplications, they would not be seen in the numbers of
observed very recent duplications from which the rate estimate is derived.
Basically, then there always are duplicates present in genomes, some of
which may be usefully converted to a slightly different functional
activity dependent upon the selective environment. What Sean needs to
show is that *all* of these duplicates are "thousands of amino acids"
away from *any possible* new function. Given the facts, that seems to
be a fool's errand.
Sean Pitman
> *Screeeech*... WARNING: goal post shift in progress! Please stand
> clear of logical contortions!
>
> Sean, it seems to me that this is exactly an example of a function that
> you claimed evolution would not be able to produce "this side of a
> zillion years". It is clearly a new type of function - *Swarm Motility*
> - that the bacterial colony has not possessed previously. It is
> unquestionably highly complex, requiring interactions of many thousands
> of amino acids in the same time - far more complex, in terms of such
> interactions, than, say, the oft-touted bacterial flagellum. It is even
> irreducible: knock out either the gene coding for A motility, or the
> gene coding for the fibril matrix, and *Swarm Motility* is broken. And
> yet, somehow, it evolved.
Many types of antibiotic resistance are based on this sort of
increased production. Penicillin resistance via an increased
production of a pre-established penicillinase enzyme is just one
example (the penicillinase code itself has never been shown to evolve
de novo). To better understand the level of complexity that we are
talking about here, consider that even the evolution of the nylonase
function was far more complex that the evolution of the swarming
function in the mutant M. xanthus bacteria. A brand new sequence with
a brand new function was required for the nylonase function to be
realized. This was not the case for the swarming function of M.
xanthus. Increasing or decreasing the production of a pre-established
gene and protein sequence, like the matrix protein in this case, is
not a problem. The problem is getting several different proteins to
come together in a unique way to do a unique function that was never
possible before. A novel function was not evolved in this case via
any sort of novel assembly of protein structure. As with a
pre-established penicillinase enzyme, the necessary structure was
already there. All that was needed is more of it via increased
production of a single gene product. As a result, there was an
increase in the amount of assembly of a previously established process
and an increased amount of swarming (Not all that complex relatively
speaking).
The question and problem for evolution is, how to acquire new
beneficial functions that require more than a simple increased
production of a particular gene product. If a simple increase in
production of a gene product explains all of the functional diversity
that we see in living things, then evolution would not be a problem to
understand at all. The fact is, many new systems of function require
much more novel assembly of different amino acid sequences. No simple
increased production of what is already there is going to achieve
this. For example, if a bacterial colony does not have a lactase
enzyme, it will not help the bacteria in this colony to make more of
the enzymes that they already have. Not at all. In order to get a
new function like the lactase function they need to evolve a new amino
acid sequence. Depending upon the rarity of such a sequence in
sequence space, the evolution of such a new sequence with a novel
function might take quite a while indeed. In the case of the lactase
function in particular, most life forms cannot evolve it even given
millions of generations of observed time.
That is the problem for you evolutionists. You need to explain how
new functions requiring new genetic structure can evolve at higher
levels of functional complexity.
This is all I have time for now . . .
> Cheers,
> RobinGoodfellow.
Sean Pitman
> > Actually the M. xanthus bacteria did not develop their extracellular
> > matrix de novo. They were in fact already able to make this matrix
> > via the use of pre-existent genes. What happened is that when Velicer
> > and Yu deleated the genes that gave these bacteria that ability to
> > achieve S-type swarming, they increased their ability to make more
> > matrix necessary for A-type swarming. What they evolved was the
> > ability to make "enhanced quantities" of a matrix which they already
> > made. In fact, genetic mutants of the evolved strains were constructed
> > by Velicer and Yu that were unable to make this fibrilar matrix even a
> > little bit, and guess what, these bacteria never could evolve any type
> > of significant swarming ability, much less the A-type or S-types.
> > "Both the genetic and chemical inhibition of fibril-matrix
> > construction inhibited the evolved swarming phenotypes."
> >
>
> In other words, this example does not disprove your strawman model of
> how complex systems evolve because it evolved by some other means than
> your strawman model.
Where is the complexity of the system that evolved here? Can you
evolve flagellar-like motility system with the increased production of
a single gene product in a historically non-motile bacterial colony?
No! You must be able to bring different proteins together in a unique
assembly that would not assemble spontaneously if the protein parts
were put together at the same time in any sort of relative
concentrations. The complexity of the flagellar system of motility
is dependent upon the actually putting together of different building
blocks in a very specific way. The addition of more of one type of
building block is not going to help until all the types of building
blocks that are required for minimum function are put in their proper
places. Having whole lot of tubulin is not going to help gain the
motility function of the flagellum if the other protein parts are not
there in sufficient quantity and proper timing.
> The wild strains could not swarm using A-type mobility. The initial
> mutants couldn't swarm *at all*.
Actually I am suggesting that the initial mutant strains could swarm
just a little bit since they did produce the necessary matrix just a
little bit, but perhaps not enough to be selectably advantageous in
their selective media. The same is true of bacteria that can produce a
little bit of penicillinase enzyme. They would be a little bit
resistant to the penicillin antibiotic, but perhaps not enough to be
very advantageous in a high concentration of penicillin until more
penicillinase enzyme is produced.
> The evolved mutants could. A novel
> swarming mobility is, by any reasonable definition, a new function,
> and it fits your description of a multi-protein function, etc.
Consider a species of blind cave fish that have lost their ability to
make eyes with just single point mutation to a particular gene. If
you take these fish out of the cave and put them in an environment
where sited eyes would be beneficial, would a sudden appearance of
eyes in the offspring of these fish be a demonstration that the level
of complexity found in fully formed fish eyes can evolve with just one
point mutation? Certainly it would not. The same thing is true here.
No new product or function evolved here. Exactly the same matrix was
produced as before, just in greater quantities. With each increase in
quantity the A-type swarming function also increased. Clearly the
swarming function is fairly complex since, without the genes for
either S- or A-type swarming ability, no other genetic sequences
evolved in these bacteria to produce the swarming effect. No new
combinations of proteins were obtained.
Clearly, if you already have the right amino acid combination, a
simple increase or decrease in production is not a complicated issue.
The problems for evolution come when correct amino acid assembly is
not initially present. If you really want to demonstrate evolution of
higher levels of complexity in some sort of convincing way, take a
bacterial colony that never had the swarming function and try to get
that type of bacterial colony to evolve this swarming function or any
other type of motility function, and see what happens.
> And it
> evolved the way I and others have said such things evolve: by gradual
> modification of a precursor structure, not by popping into being de
> novo.
There were no precursor structures here. Please try and understand
this concept. No new structures evolved. There was ONLY an
up-regulation of what was already there. Absolutely nothing new in
structure or structure function came into being here at all. That is
a problem because without the pre-existence of exactly such a matrix,
no swarming function would be realized.
> > http://www.eurekalert.org/pub_releases/2003-09/m-wsb090503.php
> >
> > Basically, this is no different than the evolution of penicillin
> > resistance in bacteria who already have the gene for penicillinase and
> > all they need to do is make more penicillinase enzyme by decreasing
> > suppression of penicillinase production. The evolution of decreased
> > suppression of production of a fairly complex enzyme like
> > penicillinase is not nearly as difficult as is the evolution of the
> > penicillinase code to begin with. A single point mutation (many
> > different ones) can decrease suppression of production. The same
> > thing happened with this case of M. xanthus swarming evolution.
>
> In a sense, you are absolutely correct that the novel swarming
> mobility is "no different" than the de-suppression of penicillinase:
> they are not fundamentally different in the evolutionary processes
> required to make them happen.
True, and that is the problem for your position. Many types of novel
functions would require much more than a simple de-suppression of
protein production. New types of proteins would be needed, entirely
new sequences evolved. Really, you must go beyond those functions
that require nothing more than an increased or decreased production of
exactly the same sequences.
> And this sense in which you are correct
> is devastating to your entire "neutral gaps" argument.
Not at all, for reasons detailed above.
This is all I have time for now. . .
> Von Smith
howard hershey
Sean Pitman wrote:
> "Ian Musgrave & Peta O'Donohue" <reynella_R...@werple.mira.net.au> wrote in message news:<1sr8tvkueebkmhchj...@4ax.com>...
>
>
>>Lets have a look at how well Dr. Pitmans ideas are supported by two
>>big proteins, trabeculin-alpha (5938 aa) and the aptly named titin
>>(27118 aa). These proteins help refute what remains of Dr. Pitmans
>>thesis.
>
>
> Oh, this should be good!
>
>
>>Firstly, some background, as more and more novel functional proteins
>>have been shown to be produced by mutation and selection, and as the
>>known density of functional proteins in protein space have dealt a
>>savage blow to Dr. Pitmans concept of neutral gaps, he has since
>>re-defined his "neutral gap" theory to apply to proteins of several
>>thousand amino acids long.
>
>
> The neutral gap problem has always been about the density of
> beneficial sequences in sequence space.
And the only relevant thing is how far the starting sequence is from
*any* beneficial sequence. Your argument claims that the starting
sequence is always "thousands" of changes away and claims that there can
be no function but the teleological or end function.
That is a strawman argument with precisely zero relationship to the real
models of how such systems arise.
> You haven't yet provided me
> with any reasonable support for your notion that the density of
> beneficial sequences is as high as 10e11 for the 100aa level for
> fairly specified proteins (such as cytochrome c and the like) and that
> it stays at this density at increasing levels of minimum amino acid
> requirements. For example, when I asked you how many lactase
> sequences exist in sequence space, you gave a number of around 10e90.
> The problem here is that the lactase function seems to require more
> than 400aa in a fairly specified order.
No it doesn't. The *function* of most enzymes requires a small number
of very specific amino acids at specific sites in very specific 3-D
arrangements for *optimal* activity. Most of the amino acids in a
protein or system do not need to be *specific*, but only *generic* in
similarity (with greater or lesser tolerance wrt *numbers* of amino
acids that are not similar). A significant number of sites only need to
*not* be specific amino acids. And, of course, real evolutionary models
do not start with proteins with a random sequence wrt the positioning
and sequence of amino acids. These proteins start with sequences,
usually, where the positioning of relevant amino acids is present, and
often where only some of the "small number of very specific amino acids
at specific sites" need to be changed to modify function. In fact,
given that most enzymes have both primary substrates and functions and
secondary substrates and functions -- that is, they bind any of a large
number of related compounds with different affinities and can also
generate related but different results -- the usual process involves
only making those changes which change the relative affinity for
secondary vrs primary substrate. That can easily be accomplished only a
few or even one amino acid change.
Why? Those amino acids must be in a fairly specified order.
> Also, the level of
> complexity is determined by the *minimum* amino acid requirement
> needed to give rise to a particular type of function - not the maximum
> amino acid system which can be found to have a particular function.
Given that the *function* depends upon the long stretch of duplications,
where is the point where the *function* cease to be observed? One
unduplicated subunit? 20? 100?
> For example, E.coli bacteria usually use a lactase gene that codes for
> an amino acid sequence of around 1000aa. This subunit sequence
> combines with 3 other identical subunits to form the lactase enzyme
> with 4000aa total. However, this is not, as you have pointed out so
> clearly yourself, the minimum amino acid requirement needed to achieve
> the lactase function to at least some selectably beneficial level. The
> same thing is true of the evolved "ebg"-lactase enzyme who's gene
> produces a subunit protein of around 1000aa that forms a hexamer with
> 5 other identical copies for a total of around 6000aa. This is not an
> example of evolution at the level of 6000aa since the gene only codes
> for 1000 of them and of these, only 480 or so seem to be absolutely
> required in a fairly specified order for minimum lactase function to
> be realized.
>
> Now, do you understand better what my criteria are for sequence
> complexity?
Thus you are specifically ruling out nearly all existing single proteins
(remember that only 0.2% of peptides are 5000 amino acids or larger and
nearly all of these you exclude for other reasons), right? There seem
to be no single proteins in living organisms which actually meet your
criteria for being unevolvable. Exactly how does the fact that there
are no real proteins which meet your requirements for being unevolvable
support your argument that proteins in cells are unevolvable?
>>To evaluate this, lets look at the nature of proteins that are 5,000
>>aa's long (using the 5000 aa length form his statement above). Are
>>the novel, functionally complex proteins which show the thumb-print of
>>the Intelligent Designer hiding in the >5,000 aa peptides?
>>
>>The first thing to note is that peptides of 5000 aa and greater
>>comprise around 0.2% of all currently known peptides, so they are
>>pretty rare. The second thing to note is that they do not comprise
>>peptides of great functional complexity; they seem to fall into three
>>broad categories
>
>
> What you evidently fail to realize is that I'm not limiting myself to
> single protein sequences at all.
Indeed, it seems that there are no examples wrt single proteins that
meet your criteria.
> Such huge single protein sequences
> are most certainly rare, but multiprotein functions at this level of
> complexity are not rare,
Such as? Compared to the multiprotein functions that make up
biochemical pathways? Regulatory cascades? Remember to exclude all the
parts that involve internal duplications, obvious duplications and minor
divergences of other proteins in (or out of) the system, and all the
amino acids whose sequence seems to be irrelevant to function in each of
the examples you give. Seems to me that you are applying different
standards for single proteins and multiprotein complexes.
> and they are far more specified in their
> sequence structure than are those relatively few examples of mammoth
> single proteins - most of which require far fewer amino acids at
> minimum to have some selectably beneficial function of the same type.
Evidence? Or assertion? Only his hairdresser knows for sure.
But, of course, these smaller proteins also come from other proteins
that have sequence similarity and perform functions (but not motility).
Moreover, the only function that Sean considers (the teleologic end
point in his mind) does not preclude subsets of these proteins from
performing different, but useful, functions in cells.
>>I'm not sure Dr.
>>Pitman would want to claim Polyketide synthase as the fingerprint of
>>the designer, as it is a) simple and b) its sequence can vary up to
>>around 100 fold, hardly a highly constrained enzyme.
>
>
> It depends upon what you mean by "100-fold" variability. For example,
> the cytochrome c sequence is considered to be highly constrained and
> yet you quote Yockey as suggesting that this ~100aa sequence can vary
> between 10e60 to 10e90 sequences in a sequence space that is around
> 10e130 sequences in size. How many fold is that?
>
>
>>So lets have a look at titin and trabeculin-alpha.
>>
>>The heart muscle cell contractile system contains the thin and thick
>>filaments, and a third filament system composed of titin. Titin is the
>>largest protein known, with a single polypeptide containing 27118
>>amino acid residues. Titin constitutes ~10% of the contractile
>>filament protein mass of the heart muscle, making it the third-most
>>abundant muscle protein (after myosin and actin).
>
>
> Certainly true . . .
>
>
>>Now, titin has a reasonably simple structure, it is basically made of
>>three domains, a PEVK domain, an immunoglobulin domain and an N2B
>>domain. These are found in many different proteins. The major thing
>>here is that the immunoglobulin domain is repeated many, many times.
>
>
> Exactly. And this lowers the specificity of the amino acids in the
> titin protein significantly.
How does it lower specificity? Can the amino acids change at will? Or
is it that one has mechanisms for generating this sequence that does not
involve the "thousands" of point mutations before 'utility' is reached
that is part of Sean's strawman?
Duplication, of course, is the engine that powers much of evolution.
You say that we should ignore internal duplications (even if there are
some changes in sequence in individual duplicates), but hold a different
standard wrt whole gene duplication. In that case, the duplicated gene
that is proposed as the source of a new 'function' must be the
equivalent to a random sequence with the only way that it can acquire a
new function being to have essentially every amino acid in it
individually change to a new amino acid. Your estimates of the amount
of sequence space that *must* be crossed is grossly exagerrated if one
can merely modify a few sequences in the duplicated gene to acquire some
selectable new function.
>>To summarize, Dr. Pitmans contention that large proteins occupy parts
>>of protein space where mutation and natural selection cannot reach is
>>falsified by the nature of the known large proteins.
>
>
> Perhaps the strawman that you have built to represent my position has
> been falsified by these examples, but my actual position has not been
> falsified at all.
Your real position is that because there are no single proteins that
meet the assumptions you require for non-evolvability (because they
include internal duplicates or are the result of fusions or only
function after protease cleavage), that therefore any imaginary large
proteins that would, in your imagination, meet all your requirements
(namely that it start out from a random sequence entirely unrelated to
the end sequence) would be unevolvable.
Of course, the dismissal criteria change when, instead of internal
duplications of sequence, you have duplications of entire genes. In
that case, you automatically assume that every protein of a complex
multi-protein system arose from your hypothetical completely random
sequence, thus requiring "thousands" of changes.
All perfectly clear to everyone but you.
> Do you really think that I was unaware of these
> examples? Come on now, there are many many more examples of large
> proteins composed of polymer-like repeats of relatively few types of
> amino acid sequences/structures. These are not examples of high
> complexity at all and my original challenge took these into
> consideration.
>
>
>>These proteins
>>are almost all fusion/repeat proteins, and look exactly what we would
>>expect large proteins to look like from known mechanisms of gene
>>mutation. There is no thumbprint of an Intelligent Designer here.
>
>
> I just knew that someone would try to float something like this, but I
> didn't think it would be you. I thought more of you in particular.
> Oh well, I still think that you have it more together than most who
> frequent this forum.
It is so damn hard to find *any* single protein that meets your criteria
without being disqualified. But if you disqualify large proteins
because they include sequences which are duplicates, why do you not
disqualify smaller proteins which arose by duplication and divergence?
And if you were to apply your standards consistently, how many of the
"thousands" of amino acids in multiprotein systems would really be relevant?
>>Cheers! Ian
>
>
> Sean
> www.naturalselection.0catch.com
>
Sean Pitman
> > matrix de novo. They were in fact already able to make this matrix
> > via the use of pre-existent genes. What happened is that when Velicer
> > and Yu deleated the genes that gave these bacteria that ability to
> > achieve S-type swarming, they increased their ability to make more
> > matrix necessary for A-type swarming. What they evolved was the
> > ability to make "enhanced quantities" of a matrix which they already
> > made. In fact, genetic mutants of the evolved strains were constructed
> > by Velicer and Yu that were unable to make this fibrilar matrix even a
> > little bit, and guess what, these bacteria never could evolve any type
> > of significant swarming ability, much less the A-type or S-types.
> > "Both the genetic and chemical inhibition of fibril-matrix
> > construction inhibited the evolved swarming phenotypes."
> >
>
> In other words, this example does not disprove your strawman model of
> how complex systems evolve because it evolved by some other means than
> your strawman model.
Where is the complexity of the system that evolved here? Can you
evolve flagellar-like motility system with the increased production of
a single gene product? No! You must be able to bring different
proteins together in a unique assembly that would not assemble
spontaneously if the protein parts were put together at the same time
in any sort of relatively concentrations. The complexity of the
flagellar system of motility is dependent upon the actually putting
together of different building blocks in a very specific way. The
addition of more of one type of building block is not going to help
until all the types of building blocks that are required for minimum
function are put in their proper places. Having whole lot of tubulin
is not going to help gain the motility function of the flagellum if
the other protein parts are not there in sufficient quantity and
proper timing.
> The wild strains could not swarm using A-type mobility. The initial
> mutants couldn't swarm *at all*.
Actually I am suggesting that the initial mutant strains could swarm
just a little bit since they did produce the necessary matrix just a
little bit, but perhaps not enough to be selectably advantageous in
their selective media. The same is true of bacteria that can produce a
little bit of penicillinase enzyme. They would be a little bit
resistant to the penicillin antibiotic, but perhaps not enough to be
very advantageous in a high concentration of penicillin until more
penicillinase enzyme is produced.
> The evolved mutants could. A novel
> swarming mobility is, by any reasonable definition, a new function,
> and it fits your description of a multi-protein function, etc.
Consider a species of blind cave fish that have lost their ability to
make eyes with just single point mutation to a particular gene. If
you take these fish out of the cave and put them in an environment
where sited eyes would be beneficial, would a sudden appearance of
eyes in the offspring of these fish be a demonstration that the level
of complexity found in fully formed fish eyes can evolve with just one
point mutation? Certainly it would not. The same thing is true here.
No new product or function evolved here. Exactly the same matrix was
produced as before, just in greater quantities. With each increase in
quantity the A-type swarming function also increased. Clearly the
swarming function is fairly complex since, without the genes for
either S- or A-type swarming ability, no other genetic sequences
evolved in these bacteria to produce the swarming effect. No new
combinations of proteins were obtained.
Clearly, if you already have the right amino acid combination, a
simple increase or decrease in production is not a complicated issue.
The problems for evolution come when correct amino acid assembly is
not initially present. If you really want to demonstrate evolution of
higher levels of complexity in some sort of convincing way, take a
bacterial colony that never had the swarming function and try to get
that type of bacterial colony to evolve this swarming function or any
other type of motility function, and see what happens.
> And it
> evolved the way I and others have said such things evolve: by gradual
> modification of a precursor structure, not by popping into being de
> novo.
There were no precursor structures here. Please try and understand
this concept. No new structures evolved. There was ONLY an
up-regulation of what was already there. Absolutely nothing new in
structure or structure function came into being here at all. That is
a problem because without the pre-existence of exactly such a matrix,
no swarming function would be realized.
> > http://www.eurekalert.org/pub_releases/2003-09/m-wsb090503.php
> >
> > Basically, this is no different than the evolution of penicillin
> > resistance in bacteria who already have the gene for penicillinase and
> > all they need to do is make more penicillinase enzyme by decreasing
> > suppression of penicillinase production. The evolution of decreased
> > suppression of production of a fairly complex enzyme like
> > penicillinase is not nearly as difficult as is the evolution of the
> > penicillinase code to begin with. A single point mutation (many
> > different ones) can decrease suppression of production. The same
> > thing happened with this case of M. xanthus swarming evolution.
>
> In a sense, you are absolutely correct that the novel swarming
> mobility is "no different" than the de-suppression of penicillinase:
> they are not fundamentally different in the evolutionary processes
> required to make them happen.
True, and that is the problem for your position. Many types of novel
functions would require much more than a simply de-suppression of
protein production. New types of proteins would be needed, entirely
new sequences evolved. Really, you must go beyond those functions
that require nothing more than an increased or decreased production of
exactly the same sequences.
> And this sense in which you are correct
> is devastating to your entire "neutral gaps" argument.
Not at all, for reasons detailed above. And, this is all I have time
for now.
> Von Smith
Sean Pitman
> > matrix de novo. They were in fact already able to make this matrix
> > via the use of pre-existent genes. What happened is that when Velicer
> > and Yu deleated the genes that gave these bacteria that ability to
> > achieve S-type swarming, they increased their ability to make more
> > matrix necessary for A-type swarming. What they evolved was the
> > ability to make "enhanced quantities" of a matrix which they already
> > made. In fact, genetic mutants of the evolved strains were constructed
> > by Velicer and Yu that were unable to make this fibrilar matrix even a
> > little bit, and guess what, these bacteria never could evolve any type
> > of significant swarming ability, much less the A-type or S-types.
> > "Both the genetic and chemical inhibition of fibril-matrix
> > construction inhibited the evolved swarming phenotypes."
> >
> > http://www.eurekalert.org/pub_releases/2003-09/m-wsb090503.php
>
> *Screeeech*... WARNING: goal post shift in progress! Please stand
> clear of logical contortions!
>
> Sean, it seems to me that this is exactly an example of a function that
> you claimed evolution would not be able to produce "this side of a
> zillion years". It is clearly a new type of function - *Swarm Motility*
> of amino acids in the same time - far more complex, in terms of such
> interactions, than, say, the oft-touted bacterial flagellum. It is even
> irreducible: knock out either the gene coding for A motility, or the
> gene coding for the fibril matrix, and *Swarm Motility* is broken. And
> yet, somehow, it evolved.
It is not enough to have to right parts, you must have them in the
right place and quantity. For example, if you put all the parts for a
flagellum into solution so that they randomly interact, they will not
self-assemble to form a working flagellum. The sequential order and
concentration of part additions to solution is necessarily for proper
flagellar assembly. A simple increase in the supply of one component
will not create spontaneous self-assembly either. This is the reason
why examples of evolution that are only dependent upon the increased
production of various parts, especially single parts, are not examples
of the evolution of high complexity. Many types of antibiotic
resistance are based on this sort of increased production. Penicillin
resistance via an increased production of a pre-established
penicillinase enzyme is just one example (the penicillinase code
itself has never been shown to evolve de novo). To better understand
the level of complexity that we are talking about here, consider that
even the evolution of the nylonase function was far more complex that
the evolution of the swarming function in the mutant M. xanthus
bacteria. Increasing or decreasing the production of a
pre-established gene and protein sequences, like the matrix protein in
this case, is not a problem. The problem is getting several different
proteins to come together in a unique way to do a unique function that
was never possible before. A novel function was not evolved in this
case via any sort of novel assembly of protein structure. As with a
pre-established penicillinase enzyme, the necessary structure was
already there. All that was needed is more of it via increased
production of a single gene product. As a result, there was an
increase in the amount of assembly of a previously established process
and an increased amount of swarming (Not all that complex relatively
speaking).
This is all I have time for now . . .
> Cheers,
> RobinGoodfellow.
Deaddog
>even the evolution of the nylonase
> function was far more complex that the evolution of the swarming
> function in the mutant M. xanthus bacteria. A brand new sequence with
> a brand new function was required for the nylonase function to be
> realized.
It must really, really suck to be you, Sean. The 'impossibility' of the
evolution of nylon hydrolase is immediately contradicted by even the
briefest of literature searches:
Prijambada et al. (1995). Emergence of nylon oligomer degradation enzymes
in Pseudomonas aeruginosa PAO through experimental evolution. Appl Env
Micro 61:2020-2022.
The title sort of says it all, but to make it blazingly clear for Sean and
his friend Floppy: (a) The authors took a strain that had no apparent
propensity for nylon degradation. (b) A population of cells was grown in
the presence of the monomer comprising nylon (6-aminohexanoate, Ahx). (c)
Evolutionary variants that could utilize Ahx appeared at a surprisingly high
rate (1:10^3). (d) These evolutionary variants were further grown on the
linear dimer of Ahx, known as Ald. After a good long time (3 freaking
months) "the turbidity of the culture broth reached 1 (A600)."
Biochemical assays revealed that the strains were true derivatives of the
original Pseudomonas and that a new enzyme activity that could hydrolyze
nylon oligomers had evolved.
Don't you get it yet, Sean? Evolution is dead easy. We can evolve new
enzymes, new regulatory pathways, and new metabolic pathways with
frightening ease, and are quickly working towards the evolution of novel
organisms with genome-wide changes in function. Why do you think industry
invests heavily in directed evolution as opposed to "Intelligent Design"
principles (whatever they may be; not to be confused with computational or
rational design). It's because it works. And "Intelligent Design" does
not, unless you're trying to make strawmen like your bud Floppy there, in
which case it does an excellent job.
I think one of the more telling comments I've ever heard in the
evolution:Creation 'debate' came from an oil industry executive. I think
this guy and I would likely have disagreed on just about every issue you
could imagine, except for one. We were standing about at yet another "try
to keep Texas from being yet more of a laughing stock" meeting, and he
leaned over and growled, "Kee-rist, if this [ID] crap could make us money
don't you think we'd use it? I don't care if I find oil by appealing to a
Noachian flood, Intelligent Design, or dousing, but the fact remains that
understanding the fossil record is a goddam important thing for a geologist,
and the fossil record is the evolutionary record."
Non-woof
howard hershey
very important in medicine, which evolves *extremely* rapidly by a
process that, like evolution at the organismal level, includes random
mutation and shows the effect of strong selection for different
variants. But this protein doesn't seem to have the problems you seem to
think it should have wrt the idea that this type of evolution could not
happen in even billions of years, Sean.
Now, admittedly, this protein's two peptides are only 454 and 106 aa
long (although they do form a tetramer of 2 each of the larger and
smaller peptides and, for the form I am considering, these tetramers
form a higher order pentamer with the addition of a third protein of 136
aa being associated with this process), making, by your standards, a
total length of 696 aa for this complex self-assembling structure (10
molecules each of the large and small peptides, but counted only once,
plus the third peptide).
The most important and interesting (for my purpose, because it shows the
power of combinatorial rearrangements of different 'active sites' and
belies the notion of function being unevolvable because the sequence
space that must be covered is so large) functional part of this protein
involves a cleft which is composed of and works by an interaction
between both the largest and smallest peptides. So that cleft clearly
meets your requirement that the separate parts must interact in a
specified spatial way in order to produce a function.
Like most enzymes and other proteins, which also involve active sites
that interact with substrates or other peptides, this cleft region also
acts by interacting with and binding small molecules (but usually when
they are bound to larger molecules) or small sections of large molecules.
But unlike most proteins, which have evolved to bind a specific
substrate, this cleft is intended to produce, by a combination of random
rearrangement of structures and random mutation followed by selection,
sites able to bind *any* possible structure an organism might run into,
even those structures which did not exist in the past.
Now your argument is that random mutation and selection cannot produce
'new function' in even trillions of years because of the impossibly
large amount of sequence space that must be covered before one can hit a
sequence which has selectable affinity for a substrate if you start with
a sequence that doesn't have affinity for that substrate. Yet here we
have a protein that does just that rather (but like all random
processes, not always) successfully, not in trillions of years, but
even, for some species, in a matter of less than a month! Something is
off in your calculations. The protein system in question is at least
modestly large (almost 700 aa), yet it seems that it doesn't have to
change most of those amino acids to acheive this 'new' ability to bind,
not really tiny molecules, but larger areas of bigger molecules, with
remarkably high specificity.
Note that binding to other proteins to form multi-protein complexes is
one of the things that the cleft region can do.
Moreover, there are even examples of these molecules gaining a
selected-for *enzymatic* activity! And this is not as simple as
changing from one substrate to a closely related one. These molecules
can mutate (both by point mutation and by chimera formation) aa on
*both* peptides so radically that one variant can bind a protein,
another a polysaccharide, and yet another, nucleotides! How is this
possible within our lifetime if this is due to a random walk across
neutral gaps, Sean?
I am having real serious problems trying to fit your ideas about numbers
of amino acids required to evolve a new function (new binding ability)
and the fact that that doesn't seem to be the number that is involved in
new binding activities for this protein, which really *does* start,
relative to the end structure in the cleft, from a 'randomly generated'
sequence.
By now, I am sure that you recognize that I am talking about IgM.
Sean Pitman
> > even the evolution of the nylonase
> > function was far more complex that the evolution of the swarming
> > function in the mutant M. xanthus bacteria. A brand new sequence with
> > a brand new function was required for the nylonase function to be
> > realized.
>
> It must really, really suck to be you, Sean. The 'impossibility' of the
> evolution of nylon hydrolase is immediately contradicted by even the
> briefest of literature searches:
Who said that the nylonase function was impossible to evolve? The
nylonase function is not only simple enough to evolve, it has been
shown to evolve in real life several times. However, the nylonase
function is a relatively simple function that requires now more than a
few hundred amino acids at minimum (less than 400aa and probably
fewer). The density of nylonase sequences in this size of sequence
space is most likely quite high. The nylonase stepping stones are
relatively close together. That is why it is possible to evolve such
a simple function as nylonase.
Now, as simple as the nylonase function is to evolve, the swarming
function of M. xanthus was vastly easier to evolve since no new amino
acids sequences were required. Exactly the same genetic products were
used, but in greater quantities. The increased production of a given
gene product is much much easier to achieve than the evolution of a
brand new sequence with a brand new function - even if it be a
relatively simple function like the nylonase, lactase, or other such
single protein enzymes or multiprotein complexes that require, at
minimum, no more than a few hundred amino acids working together at
the same time in a relatively specified order.
When you move up the ladder of complexity to novel functions that
require new genetic sequences that code for functions requiring a few
thousand amino acids working at the same time (at minimum), then you
run into real problems. No evolution happens at such levels of
functional complexity because the amount and ordered specificity
required by many functions at such levels makes these beneficial
functions extremely rare in sequence space. Finding a novel function
with a new sequence structure at such a level of complexity involves
the crossing of literal universes of non-beneficial sequences and
trillions upon trillions of years of average time.
Ian Musgrave & Peta O'Donohue
Address altered to avoid spam, delete RemoveInsert
On Mon, 8 Dec 2003 17:52:31 +0000 (UTC),
seanpi...@naturalselection.0catch.com (Sean Pitman) wrote:
>"Ian Musgrave & Peta O'Donohue" <reynella_R...@werple.mira.net.au> wrote in message news:<1h77tv45ip3g4r69o...@4ax.com>...
>
>> >You have shown that short amino acid sequences can come together to
>> >form a new unified function that is indeed unique and of greater
>> >complexity. The only problem you have here is that the minimum amino
>> >acid number required for your most complex example is only 3 or 4
>> >hundred amino acids.
>>
>> Coincidentally the median protein size is around 200-250 aa's, so the
>> amjority of proteins are covered in those kinds of experiments.
>
>It is good to see you back Ian. Hope your interesting life calmed
>down just a bit.
No. Amongst other things I have a professor AWOL in China, and typing
with baby Andy on my lap just isn't working.
Bis spaeter
Deaddog
>Finding a novel function
> with a new sequence structure at such a level of complexity involves
> the crossing of literal universes of non-beneficial sequences and
> trillions upon trillions of years of average time.
In other words, you have no actual examples whatsoever of a function that
you would say 'can't' evolve. When you do put forward examples
(beta-galactosidase, nylon hydrolase, and so forth), and your examples are
immediately shown to be relatively simple to generate by evolution, you back
off of your examples.
You blather interminably about 'complexity,' but you seem rather afraid to
actually make a concrete statement about a complex function that absolutely,
positively cannot evolve.
Could this be because you've been proven wrong so many times, or is this
instead because you know so little that you can't do much more than parrot
tired ID arguments, only to find out that they had no validity in the first
place?
Non-woof
Sean Pitman
> By now, I am sure that you recognize that I am talking about IgM.
Antibodies, like IgM, are generally produced in the vastness of their
antigen specificities before they are "selected" by antigen
recognition for clonal expansion. They do not evolve diversity over
time via any sort of selection process like you seem to think. The
diversity is created rather quickly, in the beginning, via a very
functionally complex preprogrammed mechanism that is set in place for
the specific purpose of creating antibody diversity to recognize a
host of *potential* foreign antigens (and not "self" antigens) BEFORE
these potential foreign antigens are encountered. Once the antibody
diversity is formed then this diversity is refined via antigen
selection - not so much the other way around. Also, the antibodies
do not evolve any new job beyond the ability to detect antibodies.
They may actually have the lactase function or other such functions,
but this is not helpful to the host since a potentially useful
function or structure in the wrong place is not useful. And another
thing, the actual pentamer structure of IgM as well as its many other
uniquely "antibody" structures are not evolved. The antigen producing
cell is preprogrammed to form these uniquely antibody structures (just
as the formation of the monomer IgG structures are pre-programmed)
There are variable regions that provide IgM with the potential to
recognize different antigen epitopes, but the basic pentamer structure
of IgM is not a novel structure that evolves from non IgM producing
cells.
For example, starting with a creature that does not produce
antibodies or one that has lost this ability, that creature is not
going to evolve the ability to make antibodies. The code for antibody
production and variability must be in place ahead of time for a
creature to be able to use antibodies for immunity against foreign
invaders. Antibodies certainly don't just "evolve". For instance,
those with deficient immune systems, due to the inability to make
proper antibodies, do not evolve their immune systems and/or antibody
producing capabilities back again. Oh no. Often, the only way to
really help many of these people is to give them a new immune system
via a bone barrow transplant. None antibody producing cells or life
forms simply do not evolve the ability to produce these marvelously
effective protein structures even though it would benefit an
immunodeficient host if they did. Though only a few hundred amino
acids are required total, antibiodies are rather specific in form and
function. Not just any equivalently sized protein will have the
function of an antibody.
In short, evolutionists like to suggest that the immune system is an
example of Darwinian-style evolution in action, but it really is
nothing of the sort. New functions do not evolve via mutation and
natural selection. All the diversity of antigen recognition is made
pretty much in the beginning and then selected after antibody
specificity for antigen is in place. And, no new type of function,
beyond the normal antigen recognition function, is evolved by an
antibody in a way that can be used in a beneficial manner by the host
organism.
For further discussion of this concept see:
RobinGoodfellow
OK. It seems that in this case, we had the "right" parts in the
"right" place, but not in the "right" quantity. It appears that the
quantity bit is no problem for evolution.
> For example, if you put all the parts for a
> flagellum into solution so that they randomly interact, they will not
> self-assemble to form a working flagellum. The sequential order and
> concentration of part additions to solution is necessarily for proper
> flagellar assembly.
OK, I get you. Of course, you have already established - as amusing
as I find that conclusion of yours - that anything that any function
that requires a correct sequential order is no more complex than the
most complex one of its parts (where the complexity metric that you
are using is the function of the current phase of the moon, as far as
I can tell). Thus, even by your thesis, evolution should not have too
much difficulty getting the components into the right sequence. And
above we see that evolution doesn't seem to have that much trouble
adjusting the concentrations - at least to create rudimentary
functionality. Once such functionality is attained, and realized to
some benefit, gradual refinements can take place for higher
specificity, optimal balance of concentrations, e.t.c.
> A simple increase in the supply of one component
> will not create spontaneous self-assembly either. This is the reason
> why examples of evolution that are only dependent upon the increased
> production of various parts, especially single parts, are not examples
> of the evolution of high complexity. Many types of antibiotic
> resistance are based on this sort of increased production. Penicillin
> resistance via an increased production of a pre-established
> penicillinase enzyme is just one example (the penicillinase code
> itself has never been shown to evolve de novo). To better understand
> the level of complexity that we are talking about here, consider that
> even the evolution of the nylonase function was far more complex that
> the evolution of the swarming function in the mutant M. xanthus
> bacteria.
Here you are once again muddling the notion of "complexity". The
resulting function, *Swarm Motility*, is complex by your number
"number of amino acids" metric. Here, however, you are trying to
peddle the necessary amount of genetic change as a measure of
complexity. However, this measure is completely bogus, since it
cannot be computed in absolute terms - it will be different depending
on the starting point you pick. If you want point out a complex
system, and say that it could not have evolved, pick a complexity
measure and stick with it. And your complexity measure should be
absolute: you should be able to look at the system, and compute it
outright, without knowing or caring anything about its evolutionary
history.
Allow me to illustrate. Let us look at the happy M. xanthus, that is
about to acquire *Swarm Motility*. We will consider the "number of
amino acids needed" metric, which in this case is quite high -
remember that for *Swarm Motility*, the highly complex A-motility
needs to function, along with the proteins involved in the
extra-cellular fubril matrix, which are so far insufficient for
swarming, and are currently involved in chemotaxis (see the Nature
paper). Now, all these proteins have to be involved in swarming
behavior for the bacteria to derive selective benefit - yet that is
not possible without some amount of evolutionary change. However,
according to your argument, the "density of beneficial functions" at
this level of complexity is so unbelievably low, that no new functions
could possibly found by evolution's "random walk". That is, your
argument - not backed up by any actual data, but only by the most
simplistic, completely unrelated to reality statistics - is that the
probability of there being any function that just happens to require
the intergration of components that are already in place (the
A-motility system, and the fibril matrix) is, for all effects and
purposes zero. And yet, here we are - a novel complex functions just
pops up all of the sudden, and with a bare minimum of evolutionary
effort to boot. Methinks I see a contradiction.
> Increasing or decreasing the production of a
> pre-established gene and protein sequences, like the matrix protein in
> this case, is not a problem. The problem is getting several different
> proteins to come together in a unique way to do a unique function that
> was never possible before. A novel function was not evolved in this
> case via any sort of novel assembly of protein structure. As with a
> pre-established penicillinase enzyme, the necessary structure was
> already there. All that was needed is more of it via increased
> production of a single gene product. As a result, there was an
> increase in the amount of assembly of a previously established process
> and an increased amount of swarming (Not all that complex relatively
> speaking).
OK. I have an idea, Sean. Rather then having every poster here
introduce to a red-bearded, kilt-wearing, bag-pipe playing gentleman,
only to have you proclaim that he is no true Scotsman, why don't you
do the following:
a) Define a consistent, valid metric to evaluate the complexity of a
system. (I'd ask for it to be relevant too, but that would be too
much to hope for.) So far, I've seen you use at least three
interchangibly: the minimum number of amino acids, the minimum amount
of genetic information, and the needed amount of evolutionary change.
I am sure that even you'll agree that the first two are not equivalent
- and the third isn't even a valid measure, since it is defined
relative to a non-fixed starting point. So pick one, and stick with
it.
b) Using this metric, define some minimum complexity level that you
believe is unevolvable.
c) Give us at least a half-dozen examples of molecular systems that
you feel meet or exceed this complexity level, and explain your
reasons to the rest of us. Since you say that living organisms are
swarming with such unevolvable systems, this should really be a snap
for you.
This may seem like a lot to ask, but this isn't even a prelude to the
beginning of the amount of work that you'd need to do if you wanted to
seriously challenge evolution on scientific grounds. The going would
only get harder from there. In my opinion, this is because
biologists, bio-chemists, bio-physicists, and, most recently, we
bio-informaticists - among other numerous disciplines - have a
collected a staggering amount of data than could only be made sense of
(parsimoniously) in light of ToE - and no real data that would argue
against it. But you would not believe me. I won't speculate on your
reasons for this, but I doubt that there is anything anyone here can
say or do to change your mind. So I ask you to consider this - the
court of informed opinion is very much against you at the moment: more
so than it has ever been, and growing larger. If you are serios about
challenging evolution as a scientific paradigm, you are going to have
to work very, very, very hard. In terms of the amount of work you
must do, you (both the single you and the entire ID movement) are very
far from even beginning.
That is all for now. Perhaps one day, when I am a little less busy,
I'll submit a post analyzing your so-called statistical arguments, and
detailing some of the staggering mathematical, theoretical physics,
and computational difficulties one would need to deal with if one were
serious about building rigorous statistical models of "sequence
space". And without such models, all arguments about "beneficial
densities" and "zillions of years" are irrelevant and pointless.
> This is all I have time for now . . .
Take your time. All criticisms and jibes aside, thank you for
replying. You are involved in multiple discussions simulateneously,
and I am impressed with your willingness to answer your critics, and
the civility and thoughtfulness you've displayed in doing so.
Whatever else I may think of your thesis, I sincerely appreciate the
time and effort you have expanded in presenting it.
> > Cheers,
> > RobinGoodfellow.
>
> Sean
> www.naturalselection.0catch.com
howard hershey
Sean Pitman wrote:
> howard hershey <hers...@indiana.edu> wrote in message news:<br7f93$i9p$1...@hood.uits.indiana.edu>...
>
>
>>By now, I am sure that you recognize that I am talking about IgM.
>
>
> Antibodies, like IgM, are generally produced in the vastness of their
> antigen specificities before they are "selected" by antigen
> recognition for clonal expansion.
Yes. First we have the generation of mutation or variation without
respect to need for these mutants or variants followed by selection
among these variants for those which are locally useful. And subsequent
expansion of the useful variants. How does that differ from an
evolutionary mechanism? The only difference I see are that the random
mutation rate is very much higher in this local region of the gene and,
of course, that it is occurring in cells within an organism rather than
in an organism's genome.
The point, which you seem to have missed, is that the cleft, which is
the interaction of two different proteins, has a significant number of
amino acids which can vary. And, unlike a simple evolutionary
modification to change the affinities a chemically similar, but
previously secondary substrate, so that it becomes a primary substrate,
this variable region is responsible for binding a wide, wide possible
range of substrates. According to your argument, the possibility that
such a random mutational process would be able to produce *any* single
useful result because it would have to be on one of the rare islands of
utility in a sea of completely useless *for anything* sequences.
Perhaps, just perhaps, the vast sea of sequences are not completely
useless, but merely useful for other things than the sequence of current
local utility is, and the size of islands of sequence utility for
binding any particular binding epitope are not vanishingly small, but
findable in that ocean of sequences with other functions.
> They do not evolve diversity over
> time via any sort of selection process like you seem to think.
I thought I made myself clear. I agree. The diversity is not *caused*
by selection. The diversity is caused by *random mutational (including
modular as well as point) changes* produced without any prior
teleological knowledge of that change's potential utility. I.e., the
mutations or variants are produced randomly with respect to need.
> The
> diversity is created rather quickly, in the beginning, via a very
> functionally complex preprogrammed mechanism
That "complex preprogrammed mechanism" is designed to produce high local
mutation (random wrt need) rates (including switching modules and point
mutations). The resulting variants are not preprogrammed results
produced according to future need.
> that is set in place for
> the specific purpose of creating antibody diversity to recognize a
> host of *potential* foreign antigens (and not "self" antigens) BEFORE
> these potential foreign antigens are encountered.
No. Anti-self antibodies are and can be produced by the random
mutational processes that generate the variants useful against foreign
antigens. Immune cells that produce these randomly generated anti-self
antibodies are *usually* selected *against* early in life. However,
there *are* a number of quite common diseases (rheumatoid arthritis,
lupus) that are the result of anti-self antibodies that manage to sneak
through (often because the antigen in question is a weak antigen or not
common present at the time the selection against self-antigen cells
occurs). You are unaware of autoimmune diseases and you call yourself a
doctor?
> Once the antibody
> diversity is formed then this diversity is refined via antigen
> selection - not so much the other way around.
Random (wrt need) mutation plus selection for or against the results.
Rather a powerful mechanism for generating a whole range of molecular
binding sites, isn't it?
> Also, the antibodies
> do not evolve any new job beyond the ability to detect antibodies.
I am sure you meant "to detect antigens". Of course, the fact that
antibodies can have enzymatic activity never occurred to you. It has
even been possible to generate antibodies with betagalactosidase activity.
> They may actually have the lactase function or other such functions,
> but this is not helpful to the host since a potentially useful
> function or structure in the wrong place is not useful.
And this means that it is impossible to generate all those different
epitope binding sites by random mutation and variation exactly how? I
am not arguing that antibodies must evolve into entirely different
molecules. I am pointing out that there is something seriously wrong
with your ideas about the impossibility of changing a protein (one
immunoglobin) so that it can bind something *entirely chemically
different* than another protein (another immunoglobin in a different
cell) can by a process of random mutation or variation by pointing out
that this is exactly how immunoglobins do work: Random (wrt need)
mutation followed by selection (for locally useful antibodies that bind
foreign antigens and against self-antibodies).
> And another
> thing, the actual pentamer structure of IgM as well as its many other
> uniquely "antibody" structures are not evolved.
IgM's pentamer structure (which is the serum protein) is a consequence
of the interaction between the constant regions of the large chain and
the J-protein (the third protein I mentioned, in addition to the H and L
chains of the immunoglobin itself). IgM without and prior to the
synthesis of the J-protein is a membrane-bound antibody present as the
basic L2H2 tetramer. The change in constant regions (from IgM to IgA --
the other secreted form that uses J-protein, but produces a dimer rather
than a pentamer and is found in external secretions, IgE, IgD, or IgG)
that occurs in a particular lineage is a consequence of modular
switching of subparts of the final protein. However, not only is IgM
ontologically the first class of immunoglobins to appear in development,
it is also the first class to appear phylogenetically (it is the only
class found in sharks). IgM activates the complement system very
efficiently, thus facilitating the death of invading microorganisms.
Notice the similarity to the way that these different constant modules
were likely produced (tandem duplication of the same sort that produced
the repeated elements in the very large proteins that you dismissed as
being irrelevant because they were duplications of sequence) and to the
types of tandem duplications of entire genes such as seen in the
developmental sequence of betaglobins of hemoglobin. The only
significant difference is in the frequency of subsequent deletion of the
intervening sequences in particular cells. Very rare in the case of the
very large proteins (and usually the cause of disease, such as MD, when
it does happen), occassionally locally useful (as in heterozygosity for
beta-thalassemia due to resistance to malaria), or frequent and
regularly induced (as in the immunoglobins). In short, this represents
a harnessing of a normal mutational event (deletion) so that it is
semi-regulated and induced in a particular tissue. The similarity of
this to events that occur in transposition (the regulated induction of a
transposition event) are not surprising.
Remember that among the most important evolutionary mechanisms that have
been mentioned are duplications (both internal and whole gene) followed
by divergence and chimeric duplications (duplications where different
functional protein domains come together in a new molecule).
But focus, Sean. I am not talking about whether the entire IgM evolved
(even though it and the other constant regions did). I am talking about
how frequently one can find useful epitope binding sites by a process
that involves random (wrt need) mutation and subsequent selection of
those variants that have utility. That is the mechanism you say could
not produce useful results because any changes would have to pass
through large amounts of utterly useless sequence space to find an
impossibly rare useful variant. How come an immune system that uses
that mechanism so often comes up with locally useful variants? The V
regions of the immunoglobins are much larger than 5-6 or even 10-20
amino acids (remember it includes aa on two peptides).
> The antigen producing
> cell is preprogrammed to form these uniquely antibody structures (just
> as the formation of the monomer IgG structures are pre-programmed)
> There are variable regions that provide IgM with the potential to
> recognize different antigen epitopes,
Which is what I want you to focus on. How can one generate useful
variants in the variable region if *most* of the changes would produce
utterly useless sequences (completely useless for binding anything) and
only a highly specific sequence (perhaps only a single sequence in the
cleft) would bind an epitope? That is what you propose that the
functional regions of proteins look like. Most sequences would be
completely without any utility and only a very few specific sequences
involving a large number of amino acids would have any specificity for a
particular substrate. I think there is something wrong in your model of
how proteins work. I could tell you what is wrong, but others already
have and you won't listen. So I am pointing out the V region because it
directly contradicts your quantitative argument.
> but the basic pentamer structure
> of IgM is not a novel structure that evolves from non IgM producing
> cells.
Is this supposed to be meaningful? If so, why?
> For example, starting with a creature that does not produce
> antibodies or one that has lost this ability, that creature is not
> going to evolve the ability to make antibodies. The code for antibody
> production and variability must be in place ahead of time for a
> creature to be able to use antibodies for immunity against foreign
> invaders. Antibodies certainly don't just "evolve". For instance,
> those with deficient immune systems, due to the inability to make
> proper antibodies, do not evolve their immune systems and/or antibody
> producing capabilities back again. Oh no.
So, back to the strawman idea that evolution must somehow be able to
produce a dog born to a cat if it is true. Nowhere did I (or anyone
else) claim that if one were to eliminate the immunoglobin genes present
in current mammals that it would somehow poof back into existence in
five years. I am pointing out that the ontological development of
immunoglobin's epitope-recognizing seqeunce uses the same sort of
mechanisms (mutation without respect to need followed by selection,
deletion by transposon-like interactions to form chimeric proteins) that
are proposed for evolutionary changes. Phylogenetically, it is also
clear that duplication of the constant region functional domain followed
by divergence is a better and simpler explanation for the presence of
the A, D, E, and G constant regions than is their derivation by hundreds
of individual mutations in some random sequence of DNA or their magical
poofing into existence.
> Often, the only way to
> really help many of these people is to give them a new immune system
> via a bone barrow transplant. None antibody producing cells or life
> forms simply do not evolve the ability to produce these marvelously
> effective protein structures even though it would benefit an
> immunodeficient host if they did.
Evolution, of course, occurs to populations and not to individuals. And
I am not claiming that the changes in the V region represents that type
of evolution. I am claiming that it uses mechanisms that you assert
should not be able to produce useful epitope-binding sequences even in a
trillion years. The fact that these mechanisms can produce useful
epitope-binding sequences in a few years (at most) seems to be a problem
for your ideas about how proteins work.
> Though only a few hundred amino
> acids are required total, antibiodies are rather specific in form and
> function. Not just any equivalently sized protein will have the
> function of an antibody.
So who is claiming that? Not me. Stop presenting strawmen, Sean, and I
will stop pointing out that that is what you are doing.
> In short, evolutionists like to suggest that the immune system is an
> example of Darwinian-style evolution in action, but it really is
> nothing of the sort.
[Strawman alert, just in case there is anyone out there who is so
ignorant as to think this is what evolutionary biologists would claim.]
Where, in all the Darwinian literature do you see *anyone* claiming that
an immune system can evolve in five years or do so from any old random
sequence? References, please.
> New functions do not evolve via mutation and
> natural selection.
That mechanism is precisely how all the locally useful epitopes of the V
regions of immunoglobins did arise (via the mutation part) and become
expanded (via the selection part).
> All the diversity of antigen recognition is made
> pretty much in the beginning and then selected after antibody
> specificity for antigen is in place.
Selection can only select among variants that exist. Selection does not
*produce* the variants an organism needs. The full *diversity* of the V
region is a consequence of mutational mechanisms, not selection.
Selection (either against or for) determines which variants (produced
randomly wrt need) get amplified or get destroyed.
> And, no new type of function,
> beyond the normal antigen recognition function, is evolved by an
> antibody in a way that can be used in a beneficial manner by the host
> organism.
What an irrelevant comment. It misses the point completely.
howard hershey
Sean Pitman wrote:
> drea...@hotmail.com (Von Smith) wrote in message
> news:<8d74ec45.0312...@posting.google.com>...
[snip]
> Clearly, if you already have the right amino acid combination, a
> simple increase or decrease in production is not a complicated issue.
> The problems for evolution come when correct amino acid assembly is
> not initially present.
Well, duh. What do you think people have been telling you _ad nauseum_?
And pointing out that all the proposed evolutionary pathways of *every*
system involve the use of steps which, in no cases, involves starting
from scratch. They all involve evolution from proteins or duplicates of
proteins that *are* already close to an "amino acid assembly" that would
have selectable utility in a system including that protein and any other
previously evolved proteins (but none of the future proteins that would
be needed for the teleological end system).
> If you really want to demonstrate evolution of higher levels of
> complexity in some sort of convincing way, take a bacterial colony
> that never had the swarming function and try to get that type of
> bacterial colony to evolve this swarming function or any other type
> of motility function, and see what happens.
>> And it evolved the way I and others have said such things evolve:
>> by gradual modification of a precursor structure, not by popping
>> into being de novo.
> There were no precursor structures here. Please try and understand
> this concept. No new structures evolved. There was ONLY an
> up-regulation of what was already there. Absolutely nothing new in
> structure or structure function came into being here at all. That is
> a problem because without the pre-existence of exactly such a
> matrix, no swarming function would be realized.
You may complain all you want, but modification of existing systems is
still how evolution really works. Evolution does not work by your
strawman random walk from a random sequence.
[snip]
> True, and that is the problem for your position. Many types of novel
> functions would require much more than a simply de-suppression of
> protein production. New types of proteins would be needed, entirely
> new sequences evolved.
Strawman. Evolution does not work by producing entirely new sequences
from scratch. New proteins performing new functions can arise by
chimera formation, but that certainly does not involve generating
entirely new sequences from some random different sequence. New
proteins producing new functions can be produced by duplication and
divergence to a modified function (changing, say the primary substrate).
But that doesn't involve generating entirely new sequences from some
random different sequence either. When will you start presenting your
arguments against the real theory of evolution rather than against your
strawman version?
> Really, you must go beyond those functions
> that require nothing more than an increased or decreased production
> of exactly the same sequences.
Or modification of substrate or function? Apparently the only thing
that qualifies as evolution according to Sean is evolution that involves
the generation of a new protein by massive numbers of repeated point
mutations of some other completely unrelated sequence. It is soooooo
much easier to battle that strawman which is no part of *any*
evolutionary model than to even look at whether that is what is being
proposed by biologists.
howard hershey
[snip]
Of course, part of the problem is that Sean thinks it takes the whole
village to have a function; that is, he thinks that hundreds of very
specific amino acids must change to another set of hundreds of very
specific amino acids in order for any new function to arise, with large
numbers of the other arrangements being without any function.
Another part of the problem is that Sean cannot consider the possibility
of an independence of function of different parts of a protein. That
is, the V region of an immunoglobin has a specific function *related to*
the function of the entire molecule (namely binding to a particular
antigen). The C region of an immunoglobin also has a specific function,
which varies slightly depending upon which C region is present.
Obviously, however, in the case of immunoglobins, the *function* of the
V region is entirely independent of the function of the C region. Nor
is this indepedendence of function of different local regions unusual in
protein chemistry. OTOH, the *function* of the V region (binding an
antigen) is a co-operative result of specific regions on both the H and
L chains.
The idea that hundreds of specific amino acids are required for a
*function* like binding any epitope of an antigen is directly belied by
the immunoglobins, which directly show how very few amino acids actually
need to be changed in order for the variable region to be able to bind
to literally enormous numbers of extraordinarily different antigens.
In the 107 aa that are the variable region of the light chain and the
117 aa that are the variable region of the heavy chain, there are really
three much smaller regions (three on each chain, so six altogether) that
are *hypervariable*. Almost all the variation important to the binding
of the many, many, many different possible antigens to the variable
regions of immunoglobins is a consequence of changes in these six
hypervariable regions. In terms of numbers of hypervariable amino acids
that are in all six hypervariable regions, the sum total is in the range
of 15-50. Now keep in mind that even at the extreme end of 50 aa, this
does not mean that only one sequence can bind only one antigen nor that
all of those amino acids are involved in binding the epitope. Far from
it. Only some of these amino acids actually come in contact with any
particular epitope and only those that do would be important for a
particular antibody-antigen interaction.
What that means, very explicitly, is that one does *not* need to change
thousands of amino acids to produce binding even of the wide, wide range
of antigens that immunoglobins interact with. This basicly gives the
lie to Sean's idea that new function *always* requires such thousands of
such changes. That means that the total number of amino acids in a
protein or a protein system composed of two or more peptides that must
interact in space in a specific 3-D way (this is exactly what the H and
L chain do) to produce a *different* function than the current function
the protein has is much, much smaller than Sean seems to think is the case.
Bigdakine
>From: seanpi...@naturalselection.0catch.com (Sean Pitman)
>Date: 12/10/03 8:18 AM Hawaiian Standard Time
>Message-id: <80d0c26f.03121...@posting.google.com>
>
>"Deaddog" <elling...@yahoo.com> wrote in message
>news:<br5iev$he9$1...@geraldo.cc.utexas.edu>...
>
>> > even the evolution of the nylonase
>> > function was far more complex that the evolution of the swarming
>> > function in the mutant M. xanthus bacteria. A brand new sequence with
>> > a brand new function was required for the nylonase function to be
>> > realized.
>>
>> It must really, really suck to be you, Sean. The 'impossibility' of the
>> evolution of nylon hydrolase is immediately contradicted by even the
>> briefest of literature searches:
>
>Who said that the nylonase function was impossible to evolve? The
>nylonase function is not only simple enough to evolve, it has been
>shown to evolve in real life several times. However, the nylonase
>function is a relatively simple function that requires now more than a
>few hundred amino acids at minimum (less than 400aa and probably
>fewer). The density of nylonase sequences in this size of sequence
>space is most likely quite high. The nylonase stepping stones are
>relatively close together. That is why it is possible to evolve such
>a simple function as nylonase.
IOW, we need to be mindful of Sean's subliminal definition of complexity:
If its been shown to evolve its not complex. If it hasn't yet been shown to
evolve, its complex.
Is that about right?
Stuart
Dr. Stuart A. Weinstein
Ewa Beach Institute of Tectonics
"To err is human, but to really foul things up
requires a creationist"
Sean Pitman
> "Sean Pitman" <seanpi...@naturalselection.0catch.com> wrote in message
> news:80d0c26f.03121...@posting.google.com...:
> you would say 'can't' evolve. When you do put forward examples
> (beta-galactosidase, nylon hydrolase, and so forth), and your examples are
> immediately shown to be relatively simple to generate by evolution, you back
> off of your examples.
Where have you been? I myself in my earier posts put forward the
examples of beta-galactosidase and nylonase as novel single protein
enzymes that could and have evolved. But, my claim is and has always
been that not much very far beyond these examples of relatively low
levels of functional complexity can evolve. Over and over again I have
put forward examples of complex systems that I felt could not evolve,
like bacterial motility systems (i.e., flagellar motility systems and
the like) and have explained why such systems of function could not
evolve.
Such a multiprotein system requires, at minimum, a couple dozen
different, fairly specified, proteins working together at the same
time. Each of these proteins is made up of several hundred amino
acids, on average, working together at the same time for a combined
total of more than 5 to 10 thousand fairly specified amino acids
working together at the same time, at minimum, for such types of
multiprotein functions to be realized.
Each additional minimum amino acid requirement, for a particular
*type* of function, increases the sequence space 20 fold. The problem
here is that such an increase in the level of minimum functional
complexity does not result in a 20-fold increase in the total number
of beneficial sequences in sequence space, but only a small fraction
of this amount. So, with each increase in the level of functional
complexity, the density of beneficial sequences decreases
exponentially. This decrease in the density of beneficial sequences
creates rapidly expanding neutral gaps between islands of beneficial
sequences in sequence space. At the level of a few thousand amino
acids in fairly specified order, the beneficial sequences in such a
sequence space are simply too far apart. The finding of a new
beneficial function (requiring a new amino acid sequence within such a
level of complexity) would simply take trillions upon trillions upon
zillions of years to achieve - on average.
> You blather interminably about 'complexity,' but you seem rather afraid to
> actually make a concrete statement about a complex function that absolutely,
> positively cannot evolve.
Any novel beneficial function that requires a new sequence of a few
thousand fairly specified amino acids working together at the same
time cannot evolve since the average time needed to find such a novel
beneficial sequence at this level would take more than trillions upon
trillions of years - even given a population of all the bacteria on
Earth. All life forms do in fact have functions at such levels of
complexity and beyond. Bacterial motility systems, like the flagellar
system and various other systems of motility, are well-known examples.
The problem with you evolutionists is that the best and more complex
example of real time evolution that you can come up with requires no
more than a few hundred fairly specified amino acids working together
at the same time (i.e., single protein enzymes such as lactase or
nylonase, and a few multiprotein enzymes that still total less than a
few hundred amino acids). You really have nothing to that comes
remotely close to touching much less disproving my hypothesis.
> Could this be because you've been proven wrong so many times, or is this
> instead because you know so little that you can't do much more than parrot
> tired ID arguments, only to find out that they had no validity in the first
> place?
You have no idea what you are talking about. You try and use the same
old and tired arguments that evolutionists, such as yourself, think
are so powerful. You show that the lowest levels of functional
complexity can evolve and think that you can extrapolate this to
higher and higher levels without demonstration when the evidence
clearly shows an exponential decline in the ability of mindless
evolutionary processes to evolve novel, beneficial functions with each
step up the ladder of functional complexity (each step involves a
single amino acid increase in the minimum part requirement for a
particular type of function to be realized). You have nothing beyond
the level of a few hundred fairly specified amino acids. That is your
problem. This is the challenge that you need to start working on if
you really want your comments to be a relevant attack on my position.
But first you need to at least try and catch up in your understanding
of the issues being discussed here.
> Non-woof
howard hershey
Of the many "interesting" ideas that Sean has about biology (including
the idea that the total number of amino acids in a protein determines
its function in a strictly quantitative sense, that new proteins arise
from random starting points, and that autoimmune diseases don't exist
because the system that makes antibodies cannot make antibodies
against self) his ideas about self-assembly are perhaps the most
"interesting".
Somehow, the fact that the self-assembly of bacterial flagella, like
many of the self-assembled structures in biology, is an ordered
process (that is, it's formation follows a step-wise developmental or
ontological scheme -- certain parts self-assemble first, followed by
changes which allow the next part to self-assemble to the previously
made structure) rather than being a case of an n-body aggregation or
random any-order accumulation or aggregation of parts is evidence, to
Sean, if not to anyone else, that the system doesn't self-assemble.
Yet Sean never specifies *where* this system which cannot
self-assemble is actually made nor who is involved in their assembly,
since they cannot self-assemble.
Where, Sean, are your proposed bacterial flagella factories with their
intelligent little workers who assemble the bacterial flagella and
perhaps put them in UPS trucks to get them shipped to the bacteria?
Perhaps at the North Pole? Have the factory owners shipped the jobs
to China, where the wages and environmental protections are low?
Call me a killjoy (who says that there is no bacteria flagella Santa
Claus), but I would actually *expect* bacterial flagella to
self-assemble in an ordered mechanism, especially if it evolved in a
step-wise fashion. I would expect the step-wise ontological process
of self-assembly to resemble and reflect any proposed phylogenetic
process. And I would also expect each step in the formation of
flagella to involve self-assembly rather than involving some outside
manufacturing agent. But perhaps Sean has some actual evidence for
the non-self-assembly of some system in cells that I don't know about.
I haven't seen that evidence yet. But perhaps that is just another
"interesting" "fact" in the world according to Sean.
Ian Musgrave & Peta O'Donohue
Address altered to avoid spam, delete RemoveInsert
On Mon, 8 Dec 2003 17:52:31 +0000 (UTC),
seanpi...@naturalselection.0catch.com (Sean Pitman) wrote:
>"Ian Musgrave & Peta O'Donohue" <reynella_R...@werple.mira.net.au> wrote in message news:<1h77tv45ip3g4r69o...@4ax.com>...
>
>> >You have shown that short amino acid sequences can come together to
>> >form a new unified function that is indeed unique and of greater
>> >complexity. The only problem you have here is that the minimum amino
>> >acid number required for your most complex example is only 3 or 4
>> >hundred amino acids.
>>
>> Coincidentally the median protein size is around 200-250 aa's, so the
>> amjority of proteins are covered in those kinds of experiments.
>
>It is good to see you back Ian. Hope your interesting life calmed
>down just a bit.
Nah, still "interesting", hence my spotty replies. This one has
required two nappy changes, a bottle, a glass of milo and a
gingerbread man, all at ungodly hours of the morning.
>In any case, certainly the average single protein size is around 200
>to 300aa, but there are many functions at higher levels of complexity
>that require more than this average number of amino acids to achieve
>minimum beneficial function. Many functions require multiple proteins
>working at the same time,
The vast majority of them "cascades" which you dismiss as trivial to
evolve, like the polyphenol degradation pathway.
>or at least larger proteins with multiple
>unique domains working at the same time.
Except most of these domains are not unique, and domain recycling via
protein fusion is a major force in protein evolution. An interesting
example is chimerin, a fusion between the regulatory half of protein
kinase C and the GTPase domain of our old friend BCR. The subsequent
protein, chimerin, can regulate RAS GTPases in concert with
stimulation of G-protein coupled receptors. A new function has come
from recycling old domains. Many, many proteins are like this.
>> >My argument is that evolution becomes more and
>> >more difficult the greater the minimum amino acid requirement until it
>> >becomes impossible this side of zillions of years when the minimum
>> >requirement reaches a few thousand amino acids in fairly specified
>> >order.
>>
>> Your argument is wrong, and your "minimum requirment" is irrelevant to
>> evolutinary biology.
>
>So you say, but your claim that the total number of beneficial
>sequences that have a particular function, such as the lactase
>function, is less than 10e100 or so is a problem for you.
That was before I was fully aware of the enormous amount of
variability in the lactase sequence. It's functional space is much,
much, larger than that, but I haven't had the time to do a full
sequence comparison to get an accurate figure.
>By your own
>statements and research into this question it seems like the minimum
>amino acid requirement for beneficial lactase function in a living
>thing is greater than 400aa.
No, in _modern_ organisms the smallest lactase known so far is around
400 aa long, there is ample reason to suspect from the evolved gbg
lactase, and the beta-galactiosidases evolved from glucuronidases, and
the beta-galactosidases evolved form antibodies, that the size can be
much smaller (and don't forget the lesson of polyketides synthetase).
>The sequence space at this level is
>around 10e520. This creates a ratio of lactase vs. non-lactase
>sequences of around 1 in 10e420 - using your own numbers.
Form a rough calculation done when I was not aware of the degree of
variability in the LacZ enzyme. To get a good estimate will require a
careful alignment of all existing sequences and a careful
investigation of neutral and nearly neutral substitutions, but my off
the cuff calculation would place the number of lactase sequences in
the order of 10e400. However, I need time sitting with multiple BLAST
alignments to firm this up, time I currently don't have.
>This ratio
>is a problem since many other types of functions would probably have a
>similar ratio at this level of complexity.
As the figure is not relevant, you can't say this. Indeed, from what
we know of proteins, and how they are constructed, we would be more
confident in saying that all proteins have a high fraction of
sequences functional.
>Even if there where a
>billion different types of beneficial functions in this sequence space
More likely around one in every 10e11 sequences is functional, based
on catalytic anti-bodies, and experiments on de-novo evolution of
structual and catalytic proteins (references in earlier posts).
>and even if each one of these types of functions was coded for by
>10e100 amino acid sequences, the total number of beneficial sequences
>would still be only 10e109. The ratio of beneficial vs.
>non-beneficial would still be well over 1 in 10e400.
Irrelevant, since the figure you base this on are wrong.
>Now that is a
>problem for your position. How do you get from one type of function
>to a new type of collective amino acid function across such a gap that
>grows much much greater than this at higher and higher levels of
>minimum amino acid requirements?
One amino acid at a time. Here's a thought experiment for you. Take
and enzyme of around 200 aas, say the histidine synthesizing protein
alpha/beta barrel protein HisF, this can be converted tyrosine
synthesizing protein TyrF by one aa mutation. Now add a 200 aa protein
such as Green Fluorescent Protein to the C-terminal end of HisF. THe
combined protein is 400 aa long, and the TryF function should be lost
in the welter of sequences, but you can still get there with one
single mutation.
>> >Where are your examples of evolution requiring such a level of
>> >minimum amino acid specificity?
>>
>> Evolution doesn't "require" any such thing. Most protein function
>> revolves around between 6-12 amino acids, the rest of the protein is
>> just origami to get those amino acids close to each other. An example
>> is the serine proteases, where you can use almost any sort of scaffold
>> to get ser, his and agr close to enough each other for function.
>
>Not every protein function is as flexible as a serine protease.
No, many are far more flexible. Take Legheamoglobin and myoglobin,
which have virtually identical functions with no more amino acid
positions in common than would be expected by chance.
>Many
>relatively simple single protein functions, although requiring fewer
>amino acids at minimum, are far more constricted in their sequence
>flexibility. Cytochrome c is just one example.
Of a protein with a very flexible sequence.
>It is true that some
>positions are far more flexible than others, but many positions are
>more constricted to certain categories of amino acids (acidic, basic,
>hydrophilic, hydrophobic, etc). Many other positions are restricted
>to just a handful of amino acid options, and a few are highly
>restricted to just one or two amino acid options.
This means that roughly 10^90 of sequences the length of cytochrome C
are functionally equivalent to cytochrome C. Then ther are all the
proteins that almost, but not quite cytochrome C.
>Also, even though
>just about any amino acid can be mutated, one at a time, in a protein
>sequence without a complete loss of beneficial function, it is much
>harder to change too many proteins at the same time without a complete
>loss of beneficial function - just as it is for changing the letters
>of a paragraph written in English.
So, no model of evolution requires multiple proteins changing at the
same time.
>The point is that your statement that only 6 to 12 amino acids are
>required to be in a particular order, out of 200 to 300 amino acids,
>is a significant understatement of the true limitations that are
>required for many single protein functions to be realized. These
>limitations of the larger protein sequence create a much lower ratio
>in sequence space than would otherwise be assumed based on the
>accounting of only 6 to 12 amino acids. For example, if the ratio of
>lactase vs. non-lactase enzymes were truly based on the correct
>orientation of only 6 amino acids, the maximum ratio of lactase vs.
>non-lactase enzymes in sequence space would be as high as 1 in 64
>million. With such a high ratio of lactase sequences, an average
>bacterial colony would be able to evolve a lactase enzyme in less than
>one year. This just doesn't happen in real life. Many types of
>bacteria have not been able to evolve the relatively simple lactase
>function in 50+ years (over a million generations).
Not true. In Halls, experiment, a lactase was evolved in very short
order, after deleting THAT lactase, they didn't evolve a new lactase,
but the experimental duration was much, much less than 5 years. The
longest running experimental evolution project (lenski's work) is
less than 10 years, and they haven't even looked at lactase.
Mean while, two other groups have evolved two different types of
beta-galactosidase from glucuronidases, and lactase has been evolved
from antibodies (references in other posts).
So, lets see, 4 separate beta-galactosidease evolved in less than a
years time frame each. The experimental work supports my theoretical
estimates.
>Oh no, many
>protein functions are far more restricted than you seem to understand.
Some are highly restrict, but most aren't, not to anywhere near the
levels your oversimplified models assume.
>> >Ok - take, for example, a particular function that requires, at
>> >minimum 5,000aa at minimum to be realized.
>>
>> Name one (aside from collagen that is <evil grin>)
>
>Collagen is based on polymer assembly of short proteins in long
>strands. Obviously I'm not talking about the assembly of polymers
>here. I'm talking about minimum genetic information here.
So name one.
>Not very
>much genetic information (minimum size of gene requirement) is needed
>for collagen production - relatively speaking. For example, all
>bacterial motility systems require far more genetic information (i.e.,
>minimum base pair sequence requirement) before the motility function
>will be realized.
Yet these are _multiple_ protein systems. You cannot use the
statistics for a single contiguous peptide for collections of smaller
proteins, with their own probability spaces.
>This includes the polymer production of structures
>such as the flagellar tube etc. Even though the flagellum makes up a
>significant percentage (as far as total proteins and amino acids are
>concerned) of the total structure of the flagellar motility system,
>the minimum genetic information required to code for this system is
>what is important. The total number of different proteins required,
>not the total number of the same proteins required, is what is
>important for determining informational complexity.
And is irrelevant to your 5000aa "requirement".
>Now, I know that you have tried to suggest that bacterial motility
>systems can be produced with just two or three different types of
>proteins working together at the same time, but you have failed to
>provide an adequate demonstration or reference to support this
>statement of yours.
No, the ossicilin system is well supported, you just tried to include
everything in the bacterial cell wall as a part of the motility
system.
>The fact is that the minimum part requirement for
>types of bacterial motility like the flagellar system require more
>than 20 different proteins working together at the same time for a
>total of well over 5,000aa at minimum.
I'm assuming that you mean the eubacterial flagella (as opposed to the
archebacterial flagella, which requires around 8-11 proteins)
Again, here you are
1) conflating a multiprotein system with proteins with individual
probability spaces with a single, rigid sequence protein.
2) assuming that you need all 20 proteins turning up at the same time.
In terms of 1) the individual proteins of the 20 "required" proteins
are highly variable, and an have very divergent sequences. The key
interaction sequences are small. Some of the proteins have fused, some
split, between different flagella. So the probability spaces are
vastly different from what you imagine.
In terms of 2) consider the EPC type III virulence system of E. coli.
This is basically a flagella without MotAB. A mutation which fused
MotAB to the EPC type III virulence system would result in a flagella
of sorts. One (maybe two, depending on the MotAB/ FlagMN linkages)
mutations in one or two proteins and you have a flagella. Not all 20
popping into existence simultaneously.
Note the proteins related to MotAB are involved in secretion and
secretion based motility.
Von Smith
In spite of my previous lack of success with posting follow-ups from
google, here goes again:
> > > Actually the M. xanthus bacteria did not develop their extracellular
> > > matrix de novo. They were in fact already able to make this matrix
> > > via the use of pre-existent genes. What happened is that when Velicer
> > > and Yu deleated the genes that gave these bacteria that ability to
> > > achieve S-type swarming, they increased their ability to make more
> > > matrix necessary for A-type swarming. What they evolved was the
> > > ability to make "enhanced quantities" of a matrix which they already
> > > made. In fact, genetic mutants of the evolved strains were constructed
> > > by Velicer and Yu that were unable to make this fibrilar matrix even a
> > > little bit, and guess what, these bacteria never could evolve any type
> > > of significant swarming ability, much less the A-type or S-types.
> > > "Both the genetic and chemical inhibition of fibril-matrix
> > > construction inhibited the evolved swarming phenotypes."
> > >
> >
> > In other words, this example does not disprove your strawman model of
> > how complex systems evolve because it evolved by some other means than
> > your strawman model.
>
> Where is the complexity of the system that evolved here? Can you
> evolve flagellar-like motility system with the increased production of
> a single gene product in a historically non-motile bacterial colony?
> No!
And yet, here it is: an example of a novel swarming motility with no
more genetic change than you describe. I also notice that Ian
Musgrave has pointed out an example of at least one TTSS that may be
no more than a mutation or two away from a functioning flagellum.
> You must be able to bring different proteins together in a unique
> assembly that would not assemble spontaneously if the protein parts
> were put together at the same time in any sort of relative
> concentrations. The complexity of the flagellar system of motility
> is dependent upon the actually putting together of different building
> blocks in a very specific way. The addition of more of one type of
> building block is not going to help until all the types of building
> blocks that are required for minimum function are put in their proper
> places. Having whole lot of tubulin is not going to help gain the
> motility function of the flagellum if the other protein parts are not
> there in sufficient quantity and proper timing.
And there are perfectly viable structures that have most if not all of
these protein parts in place in precisely this fashion. It is not
clear to me where you think the insurmountable obstacle is.
>
> > The wild strains could not swarm using A-type mobility. The initial
> > mutants couldn't swarm *at all*.
>
> Actually I am suggesting that the initial mutant strains could swarm
> just a little bit since they did produce the necessary matrix just a
> little bit, but perhaps not enough to be selectably advantageous in
> their selective media. The same is true of bacteria that can produce a
> little bit of penicillinase enzyme. They would be a little bit
> resistant to the penicillin antibiotic, but perhaps not enough to be
> very advantageous in a high concentration of penicillin until more
> penicillinase enzyme is produced.
The article I saw mentions no significant swarming in the mutant
strain. It is
something that would have been hard to miss were it happening (the
resulting biofilm is visible to the naked eye).
>
> > The evolved mutants could. A novel
> > swarming mobility is, by any reasonable definition, a new function,
> > and it fits your description of a multi-protein function, etc.
>
> Consider a species of blind cave fish that have lost their ability to
> make eyes with just single point mutation to a particular gene. If
> you take these fish out of the cave and put them in an environment
> where sited eyes would be beneficial, would a sudden appearance of
> eyes in the offspring of these fish be a demonstration that the level
> of complexity found in fully formed fish eyes can evolve with just one
> point mutation? Certainly it would not.
Why not? Are you now saying that the complexity of a function is not
a property of the function itself, but rather of the evolutionary
pathway that led up to it? You have two structurally and
functionally indistinguishable eyes, but according to your argument
above, one is complex and the other is not, based presumably on the
amount of "new" genetic information required to evolve it.
But if that is the case, how does one estimate the complexity of a
function without knowledge of its precursors? Using such a metric,
simply claiming that something is
complex would be assuming your conclusion: if it evolved by gradual,
slight improvement of precursor structures, the function is not
complex, regardless of what it looks like.
I have asked you several times to demonstrate the neutral gaps between
a flagellum and any likely evolutionary precursors more than once, and
each time you have answered with something like: "You're the
evolutionist; that's your job." But here you are presenting an
argument that shoulders you with precisely that burden: the claim
that something is too complex to evolve is meaningless unless you can
tell me what it is too complex to evolve *from*.
> The same thing is true here.
> No new product or function evolved here. Exactly the same matrix was
> produced as before, just in greater quantities. With each increase in
> quantity the A-type swarming function also increased. Clearly the
> swarming function is fairly complex since, without the genes for
> either S- or A-type swarming ability, no other genetic sequences
> evolved in these bacteria to produce the swarming effect. No new
> combinations of proteins were obtained.
>
> Clearly, if you already have the right amino acid combination, a
> simple increase or decrease in production is not a complicated issue.
> The problems for evolution come when correct amino acid assembly is
> not initially present. If you really want to demonstrate evolution of
> higher levels of complexity in some sort of convincing way, take a
> bacterial colony that never had the swarming function and try to get
> that type of bacterial colony to evolve this swarming function or any
> other type of motility function, and see what happens.
That is precisely what this experiment did, and we have already seen
what can happen. The mutant strain in question did not have any
swarming motility function at all, and yet it evolved one with only a
trivial change to its genetic makeup.
>
> > And it
> > evolved the way I and others have said such things evolve: by gradual
> > modification of a precursor structure, not by popping into being de
> > novo.
>
> There were no precursor structures here. Please try and understand
> this concept. No new structures evolved.
Do you not understand what the word "precursor" means? I understand
perfectly that no new structure evolved. The function was evolved by
modifying structures that were already there, but which weren't
previously able to perform the task in question (i.e., from precursor
structures). I am starting to think you are not reading or thinking
very carefully.
> There was ONLY an
> up-regulation of what was already there. Absolutely nothing new in
> structure or structure function came into being here at all. That is
> a problem because without the pre-existence of exactly such a matrix,
> no swarming function would be realized.
>
SFW? *I'm* not the one trying to argue that evolution zaps new
proteins and structures into existence out of the blue; *you* are
doing that -over my objections and those of several other posters who
keep trying to explain to you that evolution doesn't work that way.
> > > http://www.eurekalert.org/pub_releases/2003-09/m-wsb090503.php
> > >
> > > Basically, this is no different than the evolution of penicillin
> > > resistance in bacteria who already have the gene for penicillinase and
> > > all they need to do is make more penicillinase enzyme by decreasing
> > > suppression of penicillinase production. The evolution of decreased
> > > suppression of production of a fairly complex enzyme like
> > > penicillinase is not nearly as difficult as is the evolution of the
> > > penicillinase code to begin with. A single point mutation (many
> > > different ones) can decrease suppression of production. The same
> > > thing happened with this case of M. xanthus swarming evolution.
> >
> > In a sense, you are absolutely correct that the novel swarming
> > mobility is "no different" than the de-suppression of penicillinase:
> > they are not fundamentally different in the evolutionary processes
> > required to make them happen.
>
> True, and that is the problem for your position.
A case in point for my position is a problem for my position? How?
Sometimes exchanges with you read like dialogue from a bad absurdist
play:
Dr. Pitman: Complex functions cannot evolve because they require
multiple new protein sequences all appearing together in just the
right orientation.
t.o. regular: There is no need for that. One can evolve complex
functions via slight, gradual modification of pre-existing structures.
(gives examples)
Dr. Pitman: There is no complexity here. The function only required
a slight, gradual modification of pre-existing structures.
t.o. regular: That's what I just said! Just about any function you
name, to include bacterial motility, can and has been observed to
evolve that way.
Dr. Pitman: That is not sufficient. Observing the evolution of a
motility function that requires only a slight, gradual modification of
pre-existing structures does not demonstrate that motility functions
can evolve via slight, gradual modifications of pre-existing
structures. In order to demonstrate that it is possible for motility
to evolve via slight, gradual modification of pre-existing structures,
you must be able to show that motility functions can evolve via a
dozen completely novel proteins appearing quickly together.
> Many types of novel
> functions would require much more than a simple de-suppression of
> protein production. New types of proteins would be needed, entirely
> new sequences evolved. Really, you must go beyond those functions
> that require nothing more than an increased or decreased production of
> exactly the same sequences.
Proteins can change function. They can change regulation. They can
occasionally even evolve from previously meaningless sequences. You
are yourself aware of examples of every one of these things. Where is
the big problem?
Von Smith
Fortuna nimis dat multis, satis nulli.
Deaddog
"Von Smith" <drea...@hotmail.com> wrote in message
news:8d74ec45.03121...@posting.google.com...:
I'm sorry, I thought that the following kicked ass so much that it deserved
to be lifted from the depths of the fragmented reply it appeared in:
> Sometimes exchanges with you read like dialogue from a bad absurdist
> play:
>
>
> Dr. Pitman: Complex functions cannot evolve because they require
> multiple new protein sequences all appearing together in just the
> right orientation.
>
> t.o. regular: There is no need for that. One can evolve complex
> functions via slight, gradual modification of pre-existing structures.
> (gives examples)
>
> Dr. Pitman: There is no complexity here. The function only required
> a slight, gradual modification of pre-existing structures.
>
> t.o. regular: That's what I just said! Just about any function you
> name, to include bacterial motility, can and has been observed to
> evolve that way.
>
> Dr. Pitman: That is not sufficient. Observing the evolution of a
> motility function that requires only a slight, gradual modification of
> pre-existing structures does not demonstrate that motility functions
> can evolve via slight, gradual modifications of pre-existing
> structures. In order to demonstrate that it is possible for motility
> to evolve via slight, gradual modification of pre-existing structures,
> you must be able to show that motility functions can evolve via a
> dozen completely novel proteins appearing quickly together.
I get so caught up in the details, or lack thereof, that I rarely survey the
landscape and see the commonalities. Very nice.
Non-woof
Traklman
>Date: 12/14/2003 5:36 AM Central Standard Time
>Message-id: <brhia2$h56$1...@geraldo.cc.utexas.edu>
Why, thank you. And sorry it was so fragmented. I was having problems posting
followups, so this was about the third time I had to type this out (I finally
had to resort to the unprecedented step of saving it all to notepad so I
wouldn't lose it again.) The first version, the one that disappeared, was a
bit more coherent, I think.
howard hershey
Sean Pitman wrote:
> "Deaddog" <elling...@yahoo.com> wrote in message
news:<br82uu$90a$1...@geraldo.cc.utexas.edu>...
>
>>"Sean Pitman" <seanpi...@naturalselection.0catch.com> wrote in
message
>>news:80d0c26f.03121...@posting.google.com...:
>
>
>
>>In other words, you have no actual examples whatsoever of a function that
>>you would say 'can't' evolve. When you do put forward examples
>>(beta-galactosidase, nylon hydrolase, and so forth), and your
examples are
>>immediately shown to be relatively simple to generate by evolution,
you back
>>off of your examples.
>
>
> Where have you been? I myself in my earier posts put forward the
> examples of beta-galactosidase and nylonase as novel single protein
> enzymes that could and have evolved.
And you have also stated that entire interacting systems (especially
cascading systems, but also other metabolic pathways) *can* evolve
despite the large number of amino acids involved. You were not very
clear as to *exactly* why, but it had something to do with the lower
level of complexity which had something to do with the fact that the
proteins involved in these systems do not interact with each other via
non-covalent associations, but rather with the products of other
proteins. And it seems to be that this non-covalent association that
causes 'high complexity' is somehow related to the total number of amino
acids in all the proteins in some way that is not exactly clear.
Perhaps you could explain it? You keep tossing out the word
"complexity" without telling us how it is determined quantitatively.
And by saying that there are quantitative differeneces in 'complexity',
you are saying that there is a metric by which 'complexity' can be
measured. What is it? Is it the number of proteins that non-covalently
associate? Or is it the number of amino acids in those proteins? Or is
it something else?
And very large single proteins seem to never fit the criteria of being
"complex" enough to avoid simple modification by one or a few mutational
events either. Mostly because they often seem to include internal
duplications, which, according to you, reduces their "complexity".
> But, my claim is and has always
> been that not much very far beyond these examples of relatively low
> levels of functional complexity can evolve. Over and over again I have
> put forward examples of complex systems that I felt could not evolve,
^^^^^^
That's just the problem. We are not interested in your feelings or
intuition. We are interested in your evidence and not your
unsubstantiated claims or assertions.
> like bacterial motility systems (i.e., flagellar motility systems and
> the like) and have explained why such systems of function could not
> evolve.
Yes. Your claim is that such systems have a lot of what you call
"complexity", whatever you think that is.
>
> Such a multiprotein system requires, at minimum, a couple dozen
> different, fairly specified, proteins working together at the same
> time.
Actually very few of the proteins in the bacterial flagella do any work
at all. Most of them serve a purely structural rather than functional
purpose (e.g., the whip, the bearing shaft). Only the proteins in the
motor actually *do* anything. And those domains work to convert energy
into rotation in a step-wise fashion. The only 'work' that most of
these proteins do is to get transmitted through the shaft and to
self-assemble at the terminus. And that does not occur by the proteins
"working together at the same time". So perhaps you could start by
explaining why you think the bacterial flagella meets *your* requirements?
> Each of these proteins is made up of several hundred amino
> acids, on average, working together at the same time for a combined
> total of more than 5 to 10 thousand fairly specified amino acids
> working together at the same time, at minimum, for such types of
> multiprotein functions to be realized.
And exactly why is the total number of amino acids relevant wrt how the
system evolved? What is the process of evolution that you imagine where
these numbers are relevant? The only one I can think of is a model of
evolution whereby one starts with completely random nucleotide sequences
and must reach the teleological end point (the current bacterial
flagella) with NO POSSIBLE useful intermediate by random drift alone.
Is that the model of evolution that you think biologists are proposing?
If not, which model do you think they are proposing and why do you
think the total number of amino acids is relevant?
Specifically, present, clearly and concisely, the model of evolution
that you think cannot work to produce a bacterial flagella. What is the
starting point? How does the process work to produce the current
system? Is the current system the only possible one with function? Are
there no intermediate stages with independent selectable utility in the
model of 'evolution' which you say cannot occur?
> Each additional minimum amino acid requirement, for a particular
> *type* of function, increases the sequence space 20 fold. The problem
> here is that such an increase in the level of minimum functional
> complexity does not result in a 20-fold increase in the total number
> of beneficial sequences in sequence space, but only a small fraction
> of this amount. So, with each increase in the level of functional
> complexity, the density of beneficial sequences decreases
> exponentially.
You need to explain this in English or in mathematical terms. State the
assumptions you are making. Are you proposing a model of evolution of
the bacterial flagella which has no intermediate states of independent
utility -- that is the flagella is the teleological goal and only
possible functional part of the system?
> This decrease in the density of beneficial sequences
> creates rapidly expanding neutral gaps between islands of beneficial
> sequences in sequence space. At the level of a few thousand amino
> acids in fairly specified order, the beneficial sequences in such a
> sequence space are simply too far apart.
So, in your model of the way that evolution produced the bacterial
flagella, there have to be thousands of amino acid *changes* to generate
any useful selectable function? How does that differ from saying that
the model of evolution you are proposing is one in which you start from
a random sequence and go through a random walk in hopes of hitting the
only possible functional teleological end point -- the current sequence
in E. coli's bacterial flagella? Really. How does you model of
evolution differ from that strawman?
> The finding of a new
> beneficial function (requiring a new amino acid sequence within such a
> level of complexity) would simply take trillions upon trillions upon
> zillions of years to achieve - on average.
Yet it obviously didn't. And the evidence you have *for* an alternative
mechanism is...the god of the bacterial flagella, who is unnecessary for
enzyme cascades, large proteins with internal duplications, nylonase,
and beta-galactosidase? How did your GOG (God of the Gaps) work? Do
you also think he/she/it/they were responsible for the other systems
that did not *need* him/her/it/them?
>>You blather interminably about 'complexity,' but you seem rather
afraid to
>>actually make a concrete statement about a complex function that
absolutely,
>>positively cannot evolve.
>
>
> Any novel beneficial function that requires a new sequence of a few
> thousand fairly specified amino acids working together at the same
> time cannot evolve since the average time needed to find such a novel
> beneficial sequence at this level would take more than trillions upon
> trillions of years - even given a population of all the bacteria on
> Earth.
That would certainly be true...*if* you actually had evidence that any
system actually formed by such a mechanism. Both you and every
evolutionary biologist can readily agree that no such system ever
evolved by such a mechanism. Nor did any observed system *need* to
evolve by such a mechanism. Unfortunately, that is what you are (quite
poorly) trying to show -- that some system *had* to evolve by such a
bizarre mechanism. The models for the evolution of bacterial flagella
that evolutionary biologists present, however, (inadequate though they
may be, but far better than any alternative you have presented
*evidence* for) has little relationship to your model of thousands of
amino acids magically changing via a random walk through sequence space
with no utility whatsoever.
> All life forms do in fact have functions at such levels of
> complexity and beyond. Bacterial motility systems, like the flagellar
> system and various other systems of motility, are well-known examples.
>
> The problem with you evolutionists is that the best and more complex
> example of real time evolution that you can come up with requires no
> more than a few hundred fairly specified amino acids working together
> at the same time (i.e., single protein enzymes such as lactase or
> nylonase, and a few multiprotein enzymes that still total less than a
> few hundred amino acids). You really have nothing to that comes
> remotely close to touching much less disproving my hypothesis.
Which is? Do you actually *have* a testable hypothesis that doesn't
propose a strawman or that proposes a test of your preferred mechanism?
Do you know what an hypothesis is?
>>Could this be because you've been proven wrong so many times, or is this
>>instead because you know so little that you can't do much more than
parrot
>>tired ID arguments, only to find out that they had no validity in the
first
>>place?
>
>
> You have no idea what you are talking about. You try and use the same
> old and tired arguments that evolutionists, such as yourself, think
> are so powerful. You show that the lowest levels of functional
> complexity can evolve and think that you can extrapolate this to
> higher and higher levels without demonstration when the evidence
????????
Perhaps you confused "the evidence" and "the evidenceless claim"? At
best you have provided a mathematical extrapolation based on some silly
assumptions about what would be needed to generate a complex system if
you start from a random sequence and go through a random walk to a
teleologically determined end point. Which assumptions are explicitly
not observed to be necessary in the cases of evolutionary change we (and
you) point to. These bogus 'exponential' increases in difficulty and
random walk nonsense assumptions are only invoked when it is a system
that you want to demonstrate cannot evolve. You justify these
assumptions here because of "complexity", a quantitative term that you
can't or won't present in quantitative terms. You somehow fail to
recognize that the model of evolution proposed for more complex current
systems is no different *at each step in the process* than that you
agree occurs at what you call "low levels of complexity".
> clearly shows an exponential decline in the ability of mindless
> evolutionary processes to evolve novel, beneficial functions with each
> step up the ladder of functional complexity (each step involves a
> single amino acid increase in the minimum part requirement for a
????????
How is a *change* in amino acid an increase? Are you proposing that
amino acids are added to some random sequence? Do you know what you are
saying?
> particular type of function to be realized). You have nothing beyond
> the level of a few hundred fairly specified amino acids.
You have yet to justify that the total number of amino acids is relevant
to any evolutionary modification of a protein.
> That is your
> problem. This is the challenge that you need to start working on if
> you really want your comments to be a relevant attack on my position.
> But first you need to at least try and catch up in your understanding
> of the issues being discussed here.
Oh, I think the deceased canine understands that you are proposing a man
made of straw quite well. But to make a relevant attack on your
position, you actually have to have a specific position wrt what model
of evolution you are against. The best I can make out (given your
repeated nonsense assumptions about numbers of amino acids and random
walks with no selectable functions), you are battling a model of
'evolution' that every evolutionary biologist knows is impossible.
>
>
>>Non-woof
>
>
> Sean
> www.naturalselection.0catch.com
>
Sean Pitman
> > Where is the complexity of the system that evolved here? Can you
> > evolve flagellar-like motility system with the increased production of
> > a single gene product in a historically non-motile bacterial colony?
> > No!
>
> And yet, here it is: an example of a novel swarming motility with no
> more genetic change than you describe. I also notice that Ian
> Musgrave has pointed out an example of at least one TTSS that may be
> no more than a mutation or two away from a functioning flagellum.
If such a TTSS system were only one or two amino acid changes away
from the motility function of a flagellum, then such a gap would be
crossed in a single generation. The fact of the matter is that no
TTSS secretory system has ever been observed to evolve into a motile
flagellar system. That should tell you that your little hypothesis
that only one or two amino acid changes are needed might have just a
few problems.
> > You must be able to bring different proteins together in a unique
> > assembly that would not assemble spontaneously if the protein parts
> > were put together at the same time in any sort of relative
> > concentrations. The complexity of the flagellar system of motility
> > is dependent upon the actually putting together of different building
> > blocks in a very specific way. The addition of more of one type of
> > building block is not going to help until all the types of building
> > blocks that are required for minimum function are put in their proper
> > places. Having whole lot of tubulin is not going to help gain the
> > motility function of the flagellum if the other protein parts are not
> > there in sufficient quantity and proper timing.
>
> And there are perfectly viable structures that have most if not all of
> these protein parts in place in precisely this fashion. It is not
> clear to me where you think the insurmountable obstacle is.
If all the parts are already there in their proper specified
orientation with each other, then no, there is not a problem.
However, getting a new function to evolve, even when you have all the
right parts, but not in the right order, is another story altogether.
How do you get the right parts in the right order? All the amino
acids are there already being made by all life forms to make any
necessary protein. The problem is that without the pre-established
information system for their orderly assembly coming from the codes in
the DNA, these "right" parts do not know how to self-assemble
themselves to make much of anything beyond the lowest levels of
functional complexity. The problem is that making more of what you
already have will only go so far. If you really want to evolve very
many new types of functions, you are going to have to actually evolve
some new sequence structures beyond those that you already have. For
example, the simple making of more car parts ain't gonna get you no
car. You've gotta be able to put the parts together right before the
higher complex function of a car can be realized.
> > > The wild strains could not swarm using A-type mobility. The initial
> > > mutants couldn't swarm *at all*.
> >
> > Actually I am suggesting that the initial mutant strains could swarm
> > just a little bit since they did produce the necessary matrix just a
> > little bit, but perhaps not enough to be selectably advantageous in
> > their selective media. The same is true of bacteria that can produce a
> > little bit of penicillinase enzyme. They would be a little bit
> > resistant to the penicillin antibiotic, but perhaps not enough to be
> > very advantageous in a high concentration of penicillin until more
> > penicillinase enzyme is produced.
>
> The article I saw mentions no significant swarming in the mutant
> strain. It is something that would have been hard to miss were it
> happening (the resulting biofilm is visible to the naked eye).
Very small amounts of swarming would not be "visible to the naked eye"
over short periods of time just as very low levels of penicillin
resistance would also not be all that noticeable unless very careful
finely-tuned experiments were set up to detect such minute levels of
resistance based on very low levels of penicillinase production.
> > > The evolved mutants could. A novel
> > > swarming mobility is, by any reasonable definition, a new function,
> > > and it fits your description of a multi-protein function, etc.
> >
> > Consider a species of blind cave fish that have lost their ability to
> > make eyes with just single point mutation to a particular gene. If
> > you take these fish out of the cave and put them in an environment
> > where sited eyes would be beneficial, would a sudden appearance of
> > eyes in the offspring of these fish be a demonstration that the level
> > of complexity found in fully formed fish eyes can evolve with just one
> > point mutation? Certainly it would not.
>
> Why not? Are you now saying that the complexity of a function is not
> a property of the function itself, but rather of the evolutionary
> pathway that led up to it? You have two structurally and
> functionally indistinguishable eyes, but according to your argument
> above, one is complex and the other is not, based presumably on the
> amount of "new" genetic information required to evolve it.
No, what I am saying is that if you see the sudden evolution of
something as functionally complex as a fish eye, the most logical
answer is that the ancestors of these fish previously had sited eyes
which were since lost over time by a few point mutations that blocked
key pathways of eye production. If a change in environment makes it
selectably advantageous to have eyes again, a few point mutations are
all that are required to gain the historical use of vision back again
(and this has been shown to happen in real life). However, if you
know that certain ancestor creatures never had sited eyes that
devolved over time, then the odds that the offspring of these
creatures would ever evolve the ability to see with eye-like
structures is truly impossible this size of zillions of years.
In other words, it is very easy to evolve something back again if you
just recently lost it. However, it is quite another story to evolve a
highly complex function, de novo, which never existed in the ancestral
gene pool before.
> But if that is the case, how does one estimate the complexity of a
> function without knowledge of its precursors? Using such a metric,
> simply claiming that something is complex would be assuming
> your conclusion: if it evolved by gradual, slight improvement
> of precursor structures, the function is not complex, regardless of
> what it looks like.
Not at all. The swarming function is indeed a fairly complex
function. Though not as complex as various motility functions like
flagellar motility, it does require a reasonably large number of
fairly specified amino acids working at the same time. However, the
swarming system of function did not evolve gradually in this case just
as penicillin resistance via increased penicillinase production is not
dependent upon the de novo evolution of the penicillinase function.
> I have asked you several times to demonstrate the neutral gaps between
> a flagellum and any likely evolutionary precursors more than once, and
> each time you have answered with something like: "You're the
> evolutionist; that's your job." But here you are presenting an
> argument that shoulders you with precisely that burden: the claim
> that something is too complex to evolve is meaningless unless you can
> tell me what it is too complex to evolve *from*.
Look, I have said over and over again that the supposed evolutionary
steps between functional structures such as the TTSS and the motility
system of a flagellum are too widely spaced to be crossed by the
mindless forces of random mutation and natural selection. Ian
Musgrave and many others have detailed several supposed stepping
stones, but not even one of these steps has been demonstrated to
actually evolve in the laboratory or elsewhere in nature in real time.
Certainly if the density of these beneficial stepping stones were
really as abundant as you evolutionists claim that they are, such an
evolutionary process would be a piece of cake, quickly realized in
just a few years. The fact is that these supposed beneficial stepping
stone functions are actually universes apart in sequence space,
separated by vast oceans of non-beneficial possibilities.
> > The same thing is true here.
> > No new product or function evolved here. Exactly the same matrix was
> > produced as before, just in greater quantities. With each increase in
> > quantity the A-type swarming function also increased. Clearly the
> > swarming function is fairly complex since, without the genes for
> > either S- or A-type swarming ability, no other genetic sequences
> > evolved in these bacteria to produce the swarming effect. No new
> > combinations of proteins were obtained.
> >
> > Clearly, if you already have the right amino acid combination, a
> > simple increase or decrease in production is not a complicated issue.
> > The problems for evolution come when correct amino acid assembly is
> > not initially present. If you really want to demonstrate evolution of
> > higher levels of complexity in some sort of convincing way, take a
> > bacterial colony that never had the swarming function and try to get
> > that type of bacterial colony to evolve this swarming function or any
> > other type of motility function, and see what happens.
>
> That is precisely what this experiment did, and we have already seen
> what can happen. The mutant strain in question did not have any
> swarming motility function at all, and yet it evolved one with only a
> trivial change to its genetic makeup.
There was no "change" in the genetic make up of the required
structures needed for swarming. There was no change at all. The only
change is that which increased the production of this structure. Now
tell me, does such an increase in production of what is already there
explain all the diversity that we see in the vast array of life forms
on this planet? If so, then you might have something. However, if
certain functions of higher complexity are actually different in
structural makeup, how are these differences explained? Certainly not
via an appeal to increased production of certain parts alone!
Increased production does not explain specified arrangement.
> > > And it
> > > evolved the way I and others have said such things evolve: by gradual
> > > modification of a precursor structure, not by popping into being de
> > > novo.
> >
> > There were no precursor structures here. Please try and understand
> > this concept. No new structures evolved.
>
> Do you not understand what the word "precursor" means? I understand
> perfectly that no new structure evolved.
Then you don't understand the significance of what this means. No new
structure means no new type of function evolved that requires a *new
structure* that was not already there in the ancestry of that
organism.
> The function was evolved by
> modifying structures that were already there, but which weren't
> previously able to perform the task in question (i.e., from precursor
> structures).
Now you get it wrong again. No structural modification occurred here.
There was no "modifying of structures that were already there".
There was ONLY an increased production of the intact structure. That
is all. This is not the issue in question. Once you have the
structure necessary to perform a given type of function, even in the
smallest quantities, the upregulation of production of that structure
is easy for evolution to achieve.
> I am starting to think you are not reading or thinking
> very carefully.
Ditto . . .
> > There was ONLY an
> > up-regulation of what was already there. Absolutely nothing new in
> > structure or structure function came into being here at all. That is
> > a problem because without the pre-existence of exactly such a matrix,
> > no swarming function would be realized.
>
> SFW? *I'm* not the one trying to argue that evolution zaps new
> proteins and structures into existence out of the blue; *you* are
> doing that -over my objections and those of several other posters who
> keep trying to explain to you that evolution doesn't work that way.
Evolution must explain the existence of very different *structural*
functions. You must explain how new types of functions which are
dependent upon new types of structures can be evolved over time above
the lowest levels of functional complexity. You must be able to
explain the average time required to evolve such novel structural
functions. This is what you do not seem to be able to do. The best
you come up with are the most simple levels of functional evolution or
the up-regulation of pre-established structures?! What a crock . . .
> > > > http://www.eurekalert.org/pub_releases/2003-09/m-wsb090503.php
> > > >
> > > > Basically, this is no different than the evolution of penicillin
> > > > resistance in bacteria who already have the gene for penicillinase and
> > > > all they need to do is make more penicillinase enzyme by decreasing
> > > > suppression of penicillinase production. The evolution of decreased
> > > > suppression of production of a fairly complex enzyme like
> > > > penicillinase is not nearly as difficult as is the evolution of the
> > > > penicillinase code to begin with. A single point mutation (many
> > > > different ones) can decrease suppression of production. The same
> > > > thing happened with this case of M. xanthus swarming evolution.
> > >
> > > In a sense, you are absolutely correct that the novel swarming
> > > mobility is "no different" than the de-suppression of penicillinase:
> > > they are not fundamentally different in the evolutionary processes
> > > required to make them happen.
> >
> > True, and that is the problem for your position.
>
> A case in point for my position is a problem for my position? How?
> Sometimes exchanges with you read like dialogue from a bad absurdist
> play:
>
>
> Dr. Pitman: Complex functions cannot evolve because they require
> multiple new protein sequences all appearing together in just the
> right orientation.
>
> t.o. regular: There is no need for that. One can evolve complex
> functions via slight, gradual modification of pre-existing structures.
> (gives examples)
Gives examples were no structural modification what required for the
new function at all. Come on now. You must at least try and represent
your evidence accurately.
> Dr. Pitman: There is no complexity here. The function only required
> a slight, gradual modification of pre-existing structures.
Correction: The function in question required no structural
modification whatsoever - only increased production of what was
already there.
> t.o. regular: That's what I just said! Just about any function you
> name, to include bacterial motility, can and has been observed to
> evolve that way.
Not at all. Bacterial motility has not been observed to evolve in
that way or any other way - ever. Take a historically non-motile
colony of bacteria and then get them to evolve any form of motility.
You can't do it can you? Why not? If it is so easy and if selection
pressures are high enough, such a motility function should evolve in
short order. Or, evolve any other type of function of equal or
greater complexity (minimum specified amino acid requirement of a few
thousand) where the actual required structure is new!
> Dr. Pitman: That is not sufficient. Observing the evolution of a
> motility function that requires only a slight, gradual modification of
> pre-existing structures does not demonstrate that motility functions
> can evolve via slight, gradual modifications of pre-existing
> structures. In order to demonstrate that it is possible for motility
> to evolve via slight, gradual modification of pre-existing structures,
> you must be able to show that motility functions can evolve via a
> dozen completely novel proteins appearing quickly together.
You must at least modify pre-exiting structures to get novel types of
functions, not just make more of what you already have. Where is the
structural modification here?
> > Many types of novel
> > functions would require much more than a simple de-suppression of
> > protein production. New types of proteins would be needed, entirely
> > new sequences evolved. Really, you must go beyond those functions
> > that require nothing more than an increased or decreased production of
> > exactly the same sequences.
>
> Proteins can change function. They can change regulation. They can
> occasionally even evolve from previously meaningless sequences. You
> are yourself aware of examples of every one of these things. Where is
> the big problem?
The problem is that such changes have not been observed to create
novel types of functions with new protein structures beyond the lowest
levels of functional complexity (i.e., more than a few hundred fairly
specified amino acids working together at the same time). No series
of stepwise gradual mutations have done this as you claim. The best
you have come up with are new functions requiring only a few hundred
fairly specified amino acids at minimum or those functions that do not
require any sort of new genetic sequence at all, but simply more of
what they already have. Come on now. Is that they best you
evolutionists have to base your faith on? These demonstrations do not
explain in the least how mindless evolutionary processes can or could
have created novel functions at the high level of complexity and
diversity that we see within all living things. For you millions of
years explain everything, but really a million years is like nothing
in comparison to the average time that is actually needed to find
novel functions at such levels of complexity (a few thousand specified
amino acids at minimum - like flagellar motility etc).
In fact, if you were correct in your ideas about the relatively high
density of beneficial functions in sequence space, evolution at very
high levels of complexity should be popping into existence all around
you every day. The fact that this is not happening should give you
pause to consider that perhaps the density of beneficial functions in
sequence space does not behave quite like you have imagined. Perhaps
each additional minimum amino acid requirement really does increase
the size of sequence space by 20-fold while the number of beneficial
sequences increases by only a fraction of this amount.
> Von Smith
> Fortuna nimis dat multis, satis nulli.
Ian Musgrave & Peta O'Donohue
Address altered to avoid spam, delete RemoveInsert
On Tue, 16 Dec 2003 17:02:49 +0000 (UTC),
seanpi...@naturalselection.0catch.com (Sean Pitman) wrote:
>drea...@hotmail.com (Von Smith) wrote in message news:<8d74ec45.03121...@posting.google.com>...
>> > Where is the complexity of the system that evolved here? Can you
>> > evolve flagellar-like motility system with the increased production of
>> > a single gene product in a historically non-motile bacterial colony?
>> > No!
No one has ever tried.
>> And yet, here it is: an example of a novel swarming motility with no
>> more genetic change than you describe. I also notice that Ian
>> Musgrave has pointed out an example of at least one TTSS that may be
>> no more than a mutation or two away from a functioning flagellum.
>
>If such a TTSS system were only one or two amino acid changes away
>from the motility function of a flagellum, then such a gap would be
>crossed in a single generation. The fact of the matter is that no
>TTSS secretory system has ever been observed to evolve into a motile
>flagellar system.
The fact of the matter is that no one has ever tried. Nore has anyone
one surveyed (comprehensively or otherwise) wild bacteria to see if
this has happened. We have really looked at the motility systems of
around a couple of dozen bacteria (out of millions) in any detail, so
its not suprising we haven't seen anything.
Your "evidence" is non-existant.
Deaddog
"Sean Pitman" <seanpi...@naturalselection.0catch.com> writes much the
same old nonsense in message
news:80d0c26f.03120...@posting.google.com...:
Sadly, Sean chooses to ignore the many indicated examples of complex
functional evolution that utilize, in the aggregate, proteins encoded by
many thousands of amino acids. Like any typical Creationist, he chooses to
hide in the Gaps, where he thinks we cannot find him. When presented with
the very nice example of the evolution of Myxococcus swarming (Velicer and
Yu, Nature 425:75), he shucks and jives, and claims it is not a motility
system. Instead, he focuses on flagella and other means of mobility, since
modern biology only has excellent phylogenetic and other evidence for their
evolution, as opposed to a real-time, directed evolution experiment showing
the evolution of mobility. The fact that no one has attempted such an
experiment is a warped form of 'proof' for Sean, and he hunkers down,
troll-like, in his Gap.
That's fine. Unlike Creationists, evolutionary biologists have learned to
take on the intellectual problems relating to the origin and diversity of
life. Rather than claiming that being conceptually blind, deaf, and dumb is
the only possible outcome of naturalism, evolutionary biologists learn to
seek out proof where we can, and to build that proof into the towering
monolith that is modern biological thought.
So, we can play in Sean's gap just as well as anywhere else. It will just
take a simple question, really.
Sean, would you claim that there are no cogent examples of the modern
evolution of bacterial motility, and that there is no reasonable mechanism
for the evolution of bacterial motility?
Simple question, simply stated. It really only requires an affirmation,
since I believe it is a reasonable summary of Sean's position. Doesn't
matter, since Sean will ignore the question, or post reams of irrelevant
chatter (less likely, given his propensity to hide not only in the Gaps, but
from one or more posters here as well).
That's OK. I'll keep repeating it until we get an answer. And then we can
see whether that answer stands experimental scrutiny ... or not.
Non-woof
howard hershey
Sean Pitman wrote:
> drea...@hotmail.com (Von Smith) wrote in message
> news:<8d74ec45.03121...@posting.google.com>...
>
>
>>> Where is the complexity of the system that evolved here? Can you
>>> evolve flagellar-like motility system with the increased
>>> production of a single gene product in a historically non-motile
>>> bacterial colony? No!
>>
>> And yet, here it is: an example of a novel swarming motility with
>> no more genetic change than you describe. I also notice that Ian
>> Musgrave has pointed out an example of at least one TTSS that may
>> be no more than a mutation or two away from a functioning
>> flagellum.
>
>
> If such a TTSS system were only one or two amino acid changes away
> from the motility function of a flagellum, then such a gap would be
> crossed in a single generation.
That depends on what exact changes are needed to get the new function
and on the selective environments present. Not all evolutionarily
relevant change involves point mutations. In particular duplications
and chimeric rearrangements/duplications can be important.
And a *particular* TTSS system may indeed be one or two amino acid
changes away from a motility function, but only if these occur in a
duplicate of a gene. That may mean that one would be unlikely to get
both the duplication and the mutations in a laborartory setting.
*Other* closely related TTSS systems may not need duplication but would
be 4-6 amino acid changes away. These would be unlikely to produce
'motility' in laboratory conditions that necessarily are limited in time
and numbers relative to nature. Just as the evolution of antimalarial
resistance can occur in nature but not the lab.
However, if you can identify a single protein that would need to be
changed by a few amino acids to generate a closely related protein that
would produce motility, how does that make the bacterial flagella any
different than the type of evolution we see in what you call 'less
complex' systems?
And it also depends on the selective environment. Specifically, it may
require an environment in which the level of 'motility' produced is more
useful than any accompanying loss in the 'secretion' function or in
which the balance of 'motility' and 'secretion' is better than
'secretion' alone often enough to make 'motility' useful. Not all
bacteria are better off with motility.
Note that what this proposal does NOT depend on is the fact that the
TTSS system has 'thousands of amino acids'. That is utterly irrelevant.
The only relevant metric is the number of mutational changes needed to
produce the proper protein.
> The fact of the matter is that no TTSS secretory system has ever been
> observed to evolve into a motile flagellar system. That should tell
> you that your little hypothesis that only one or two amino acid
> changes are needed might have just a few problems.
Indeed. I mentioned a few of them above. Based on the known
assumptions of the mechanisms involved. But none of them are
insurmountable problems that make such a change impossible. And you
have to demonstrate that it is *impossible* to convert a pre-existing
precursor protein in a bacteria that *already has* a TTSS system into
the type of protein that is found when one has a motile flagella which
also has TTSS activity.
Absence of evidence is not evidence of absence. If one can point out
that the main difference between a non-motile TTSS system and a motile
flagellar system (with some TTSS activity and containing proteins
related to the TTSS system) is the presence of a particular protein
which is slightly different from one that existed in the bacteria with a
non-motile TTSS system, that, in itself, cuts the legs out from under
your argument, which requires that there must be massive amounts of
difference in order for the changes to be impossible.
>>> You must be able to bring different proteins together in a unique
>>> assembly that would not assemble spontaneously if the protein
>>> parts were put together at the same time in any sort of relative
>>> concentrations. The complexity of the flagellar system of
>>> motility is dependent upon the actually putting together of
>>> different building blocks in a very specific way. The addition
>>> of more of one type of building block is not going to help until
>>> all the types of building blocks that are required for minimum
>>> function are put in their proper places. Having whole lot of
>>> tubulin is not going to help gain the motility function of the
>>> flagellum if the other protein parts are not there in sufficient
>>> quantity and proper timing.
>>
>> And there are perfectly viable structures that have most if not all
>> of these protein parts in place in precisely this fashion. It is
>> not clear to me where you think the insurmountable obstacle is.
>
>
> If all the parts are already there in their proper specified
> orientation with each other, then no, there is not a problem.
Let's be clear rather than use vague terms like 'parts'. The 'right
parts' needed to assemble a TTSS or a flagella structure are the
translated proteins (and the right environmental conditions), not the
mRNA or the DNA. The job of the mRNA is done when the protein is
translated. The job of the DNA is done when the mRNA is transcribed.
The proteins needed to make a TTSS system are the same proteins in the
same self-assembled order that would be needed to make a flagella. They
self-assemble in the same way in both cases. What are the 'parts' of
which you speak?
> However, getting a new function to evolve, even when you have all the
> right parts, but not in the right order, is another story
> altogether. How do you get the right parts in the right order? All
> the amino acids are there already being made by all life forms to
> make any necessary protein.
How is this relevant? The right parts needed for self-assembly are the
translated proteins, not the amino acids.
> The problem is that without the pre-established information system
> for their orderly assembly coming from the codes in the DNA, these
> "right" parts do not know how to self-assemble themselves to make
> much of anything beyond the lowest levels of functional complexity.
>
How is this relevant? We are saying that the TTSS system is already
present. That means that the information for making all those proteins
are already there. We are saying that the last step in converting a
TTSS system into one with some motility function involves the
modification of a pre-existing gene that makes a protein similar to the
needed protein. We are specifically NOT saying that we start with a
random sequence of DNA and generate all the proteins by a random walk
mechanism. What do you think you are saying in the above except that we
are proposing a mechanism of evolution that differs from starting with a
random sequence of DNA and generating functions from scratch by a random
walk but rather proposing one that works by descent with and by
modification of pre-existing systems? Yes. We know perfectly well that
that is what we are proposing. Why do you keep wanting us to propose
that evolution works by starting with a random sequence and generating
function by a random walk?
> The problem is that making more of what you already have will only go
> so far. If you really want to evolve very many new types of
> functions, you are going to have to actually evolve some new sequence
> structures beyond those that you already have. For example, the
> simple making of more car parts ain't gonna get you no car. You've
> gotta be able to put the parts together right before the higher
> complex function of a car can be realized.
And TTSS does just that. Think of TTSS as the body, transmission, and
wheels of a car used to house a homeless family. The next step is to
take a motor (already used for other things in the cell) and hook it up
to modify the function of the car (which can still be used as
home-sweet-home). All you need is the right connectors. Whether adding
motility to the function of the car body is 'useful' still depends on
the local conditions.
> answer is that the ancestors of these fish previously had sighted
> eyes which were since lost over time by a few point mutations that
> blocked key pathways of eye production. If a change in environment
> makes it selectably advantageous to have eyes again, a few point
> mutations are all that are required to gain the historical use of
> vision back again (and this has been shown to happen in real life).
> However, if you know that certain ancestor creatures never had
> sighted eyes that devolved over time, then the odds that the
> offspring of these creatures would ever evolve the ability to see
> with eye-like structures is truly impossible this size of zillions of
> years.
Yet the *evidence* tells us it is not impossible for eyes to evolve in
less than zillions of years. A camera eye with lens has been estimated
to take a lot less than that amount of time (see the Nilsson and Pegler
paper), using conservative estimates of the variance in these
quantitative traits, conservative estimates of heritability, and
conservative estimates of selective advantage. It is not even
impossible for *vision* to evolve independently at least twice, since it
has. It did so by re-using only the first step of the process and
co-opting already existent pathways of neural biochemistry to transmit
the signal.
> In other words, it is very easy to evolve something back again if you
> just recently lost it. However, it is quite another story to evolve
> a highly complex function, de novo, which never existed in the
> ancestral gene pool before.
What makes you think that evolution produces "highly complex functions,
de novo" *from scratch*? Evolution produces "highly complex functions"
by stepwise modification of pre-existing functions and systems. Stop
generating a strawman, Sean.
>> But if that is the case, how does one estimate the complexity of a
>> function without knowledge of its precursors? Using such a metric,
>> simply claiming that something is complex would be assuming your
>> conclusion: if it evolved by gradual, slight improvement of
>> precursor structures, the function is not complex, regardless of
>> what it looks like.
>
>
> Not at all. The swarming function is indeed a fairly complex
> function. Though not as complex as various motility functions like
> flagellar motility, it does require a reasonably large number of
> fairly specified amino acids working at the same time. However, the
> swarming system of function did not evolve gradually in this case
> just as penicillin resistance via increased penicillinase production
> is not dependent upon the de novo evolution of the penicillinase
> function.
What you mean is the exact opposite of the fact that the "swarming
system of function did not evolve gradually". Instead you object that
that the swarming system did not "poof" into existence *from scratch*
instantaneously, but only had to be modified.
>> I have asked you several times to demonstrate the neutral gaps
>> between a flagellum and any likely evolutionary precursors more
>> than once, and each time you have answered with something like:
>> "You're the evolutionist; that's your job." But here you are
>> presenting an argument that shoulders you with precisely that
>> burden: the claim that something is too complex to evolve is
>> meaningless unless you can tell me what it is too complex to evolve
>> *from*.
>
>
> Look, I have said over and over again that the supposed evolutionary
> steps between functional structures such as the TTSS and the motility
> system of a flagellum are too widely spaced to be crossed by the
> mindless forces of random mutation and natural selection.
But they are much more closely spaced than your proposal that the entire
flagella system poof into existence from scratch all at once, which is
what you think makes it impossible. *Perhaps* the number of mutational
steps between the putative TTSS and the primitive flagellum is too
widely spaced to be crossed by the mindless forces of random mutation
and natural selection *in the time available for experimentation*.
*Perhaps* the number of mutational steps depends on the pre-existence of
a duplicate of a pre-existing gene. But the number of steps required to
produce anti-malarial resistance is larger than can be experimentally
determined, and nature did that in less than 40 years under the
appropriate selective environment.
But *you* have to demonstrate that the number of *absolutely necessary*
mutational steps between these smaller selectable stages (there may be
several selectable stages in going from a particular ancestral TTSS
system to a primitive flagellar system which can undergo further
selection for greater motility) is still too large to happen in zillions
of years. That means that *you* need to find out *how many changes* are
needed to go from some putative TTSS to a primitive flagellar function
(probably also with TTSS activity). That number *certainly* will NOT be
the total number of amino acids in the flagella that you repeatedly
spout. It will be much, much, much smaller.
> Ian Musgrave and many others have detailed several supposed stepping
> stones, but not even one of these steps has been demonstrated to
> actually evolve in the laboratory or elsewhere in nature in real
> time. Certainly if the density of these beneficial stepping stones
> were really as abundant as you evolutionists claim that they are,
> such an evolutionary process would be a piece of cake, quickly
> realized in just a few years.
Only if you think evolution works by supernatural magic and only if you
think that flagellar motility is a teleologic goal. Otherwise, one
would expect the process to take geological time if one assumed that it
worked by evolutionary mechanisms.
> The fact is that these supposed beneficial stepping stone functions
> are actually universes apart in sequence space, separated by vast
> oceans of non-beneficial possibilities.
Except that the actual mechanisms do not start with random sequences in
sequence space and require that they cross oceans of non-beneficial
possibilities. They start with sequences that are already close to a
relevant functional island in an island chain. At most, evolution
involves crossing a short shallows between island with function A and
island with related function B. And does so best at low tide.
>>> The same thing is true here. No new product or function evolved
>>> here. Exactly the same matrix was produced as before, just in
>>> greater quantities. With each increase in quantity the A-type
>>> swarming function also increased. Clearly the swarming function
>>> is fairly complex since, without the genes for either S- or
>>> A-type swarming ability, no other genetic sequences evolved in
>>> these bacteria to produce the swarming effect. No new
>>> combinations of proteins were obtained.
>>>
>>> Clearly, if you already have the right amino acid combination, a
>>> simple increase or decrease in production is not a complicated
>>> issue. The problems for evolution come when correct amino acid
>>> assembly is not initially present. If you really want to
>>> demonstrate evolution of higher levels of complexity in some sort
>>> of convincing way, take a bacterial colony that never had the
>>> swarming function and try to get that type of bacterial colony to
>>> evolve this swarming function or any other type of motility
>>> function, and see what happens.
>>
>> That is precisely what this experiment did, and we have already
>> seen what can happen. The mutant strain in question did not have
>> any swarming motility function at all, and yet it evolved one with
>> only a trivial change to its genetic makeup.
>
>
> There was no "change" in the genetic make up of the required
> structures needed for swarming. There was no change at all. The only
> change is that which increased the production of this structure.
That is, one can "change" the *function* of a protein system without
changing its *structure*. At the very least, one can "change" a minor
and unimportant secondary *function* of a protein system into a major
and selectively important *function* without changing *structure*.
Another example of this would be the various eye crystallins.
The evolutionary argument about bacterial flagella similarly is an
argument about an emerging change in *function* with only a little more
change in *structure*.
Remember that your argument is the unevidenced assertion that certain
new *functions* requires a magical poofing into existence of entirely
new *structures* of 'thousands' of amino acids all at once. The above
example points out that one can even have a dramatic change in
*function* by simple mutations that affect *regulation* rather than
change in *structure* of the proteins.
> Now tell me, does such an increase in production of what is already
> there explain all the diversity that we see in the vast array of life
> forms on this planet?
Descent with and by modification of pre-existing material can explain
all the diversity we see, including, via gene and genome duplication and
divergence, all the observed increases in complexity. The evidence also
indicates *common* descent of all current life on this planet.
> If so, then you might have something. However, if certain functions
> of higher complexity are actually different in structural makeup, how
> are these differences explained?
Changes in *structure* involve the same kinds of mutational events as
the changes in *regulation*. Even entirely new proteins can be made via
chimera formation, although, unlike point mutations, the types of
rearrangements that lead to chimera protein formation are not frequently
recurrent events.
> Certainly not via an appeal to increased production of certain parts
> alone! Increased production does not explain specified arrangement.
Mutation and selection leads to change in *function* regardless of
whether that change in *function* is due to mutation in regulation or
mutation in structure.
>>>> And it evolved the way I and others have said such things
>>>> evolve: by gradual modification of a precursor structure, not
>>>> by popping into being de novo.
>>>
>>> There were no precursor structures here. Please try and
>>> understand this concept. No new structures evolved.
Is not regulation encoded in DNA, just like structure? Did not the
changes involve change (mutation) in DNA and selection for the
phenotypic effects of that change? The mechanism that produced this new
*function* is no differernt than the mutation and selection that
produces antibiotic resistance or immunoglobins that bind wildly
different epitopes. Random mutation (wrt need) is the *only* mechanism
that has been observed to produce the variation upon which selection by
the environment acts. Not a single change in DNA or structure or
function or regulation has been observed to be due to intelligent design
by a non-human designer. Indeed, the mechanism that human designers use
to identify and change organisms for his own benefit is called
'artificial selection' for a reason -- humans are merely exploiting the
mechanism that nature itself uses.
>> Do you not understand what the word "precursor" means? I
>> understand perfectly that no new structure evolved.
>
>
> Then you don't understand the significance of what this means. No
> new structure means no new type of function evolved that requires a
> *new structure* that was not already there in the ancestry of that
> organism.
>
>
>> The function was evolved by modifying structures that were already
>> there, but which weren't previously able to perform the task in
>> question (i.e., from precursor structures).
>
>
> Now you get it wrong again. No structural modification occurred
> here. There was no "modifying of structures that were already there".
> There was ONLY an increased production of the intact structure.
> That is all. This is not the issue in question. Once you have the
> structure necessary to perform a given type of function, even in the
> smallest quantities, the upregulation of production of that structure
> is easy for evolution to achieve.
Yet your basic argument is that new function cannot arise without
thousands of amino acids being produced from scratch. In this case, new
function arose without any change in protein structure (as is also the
case with eye crystallins). Every step in evolution has to be possible
to acheive. You have yet to identify which step in any specific
evolutionary scheme is, even in principle, impossible. All you do is
mumble about the total number of amino acids, the complexity!, the
complexity!, and new functions that require something made from scratch
in 5 years.
>> I am starting to think you are not reading or thinking very
>> carefully.
>
>
> Ditto . . .
>
>
>>> There was ONLY an up-regulation of what was already there.
>>> Absolutely nothing new in structure or structure function came
>>> into being here at all. That is a problem because without the
>>> pre-existence of exactly such a matrix, no swarming function
>>> would be realized.
>>
>> SFW? *I'm* not the one trying to argue that evolution zaps new
>> proteins and structures into existence out of the blue; *you* are
>> doing that -over my objections and those of several other posters
>> who keep trying to explain to you that evolution doesn't work that
>> way.
Yep. And mumbling rants about total number of amino acids and the
complexity! the complexity! the horror! of it all.
> Evolution must explain the existence of very different *structural*
> functions. You must explain how new types of functions which are
> dependent upon new types of structures can be evolved over time above
> the lowest levels of functional complexity. You must be able to
> explain the average time required to evolve such novel structural
> functions.
The time depends upon the time that the intermediate states of
functional utility are useful before the next step causes change which
is more beneficial.
> This is what you do not seem to be able to do. The best you come up
> with are the most simple levels of functional evolution or the
> up-regulation of pre-established structures?! What a crock . . .
Evolution at the simple levels of function is all that one needs for
evolution to occur.
>
>
[snip stuff that shows that this discussion resembles a bad absurdist play]
Von Smith
> drea...@hotmail.com (Von Smith) wrote in message news:<8d74ec45.03121...@posting.google.com>...
>
> > > Where is the complexity of the system that evolved here? Can you
> > > evolve flagellar-like motility system with the increased production of
> > > a single gene product in a historically non-motile bacterial colony?
> > > No!
> >
> > And yet, here it is: an example of a novel swarming motility with no
> > more genetic change than you describe. I also notice that Ian
> > Musgrave has pointed out an example of at least one TTSS that may be
> > no more than a mutation or two away from a functioning flagellum.
>
> If such a TTSS system were only one or two amino acid changes away
> from the motility function of a flagellum, then such a gap would be
> crossed in a single generation. The fact of the matter is that no
> TTSS secretory system has ever been observed to evolve into a motile
> flagellar system. That should tell you that your little hypothesis
> that only one or two amino acid changes are needed might have just a
> few problems.
Do you think that it is possible to suppress flagellum formation with
a single point mutation or not? If so, then there exists at least one
possible organism without a flagellum that is only one or two point
mutations from one that has one, and a back mutation can presumably
restore the function. You might argue that such an organism is
unlikely to have existed as a predecessor, and you may even be right.
But the point is that, in order to assess the evolvability of
anything, you must have some model of its predecessor.
In assessing whether it is feasible for a bird's wing to have evolved,
it is important to ask: evolve from what? Under what selective
pressures and constraints? Are we proposing evolving it in a raptor,
in a clam, or in a liverwort? Until you address actual scenarios, you
are speculating in a vacuum. You like to get out of this by sneering
at such scenarios as "just-so stories", perhaps hoping that no one
will point out that when you demand a detailed reconstruction of a
complex, contingent event (such as how the flagellum evolved), that is
exactly what you are demanding.
So how many neutral gaps separate a structure like a Tsp pilus from a
flagellum? Give me your calculation. A 95% confidence interval will
do fine.
Any likely precursor organism to E. coli already has DNA that can do
this, so where is the problem? You make it sound as if I have to
re-establish protein synthesis from scratch each time a new protein
evolves. And we already know that isn't the case. As to the proteins
not knowing how to self-assemble, this is really a peculiar claim; you
basically seem to be saying that protein chemistry isn't really
chemistry. You have yet to explain to me what it is, exactly that
DNA does apart from code for polypeptides.
<snip>
>
> > > > The wild strains could not swarm using A-type mobility. The initial
> > > > mutants couldn't swarm *at all*.
> > >
> > > Actually I am suggesting that the initial mutant strains could swarm
> > > just a little bit since they did produce the necessary matrix just a
> > > little bit, but perhaps not enough to be selectably advantageous in
> > > their selective media. The same is true of bacteria that can produce a
> > > little bit of penicillinase enzyme. They would be a little bit
> > > resistant to the penicillin antibiotic, but perhaps not enough to be
> > > very advantageous in a high concentration of penicillin until more
> > > penicillinase enzyme is produced.
> >
> > The article I saw mentions no significant swarming in the mutant
> > strain. It is something that would have been hard to miss were it
> > happening (the resulting biofilm is visible to the naked eye).
>
> Very small amounts of swarming would not be "visible to the naked eye"
> over short periods of time just as very low levels of penicillin
> resistance would also not be all that noticeable unless very careful
> finely-tuned experiments were set up to detect such minute levels of
> resistance based on very low levels of penicillinase production.
In order for a colony to move together, it has to produce enough
biofilm to connect the entire colony. This would have been hard to
miss. Or perhaps you mean that they were forming small amounts of
biofilm that connected only a handful of organisms at a time? Well,
this behavior would again be fairly easy to detect if it happened with
any degree of frequency, especailly considering that this was exactly
the experimenters were looking for. Do you know if any was observed?
At any rate, I'm not really sure why you think it's important to argue
this. We already agree that the organism could already use A motility
inidividually, so they were already producing minute levels of the
biofilm. You seem to want to save the idea that no new *function* has
evolved by arguing that the organisms could already use biofilms to
achieve swarming A-motility, a claim which puts you in the position of
flatly contradicting the researchers on the basis of no evidence.
In other other words, you cannot make any meaningful statement about
how easy or difficult it will be to evolve something unless you know
what it evolved from. Which is what people have been trying to tell
you all along. No one is suggesting that an eye has to pop out of
nowhere in real time, which is what "de novo" appears to mean to you.
>
> > But if that is the case, how does one estimate the complexity of a
> > function without knowledge of its precursors? Using such a metric,
> > simply claiming that something is complex would be assuming
> > your conclusion: if it evolved by gradual, slight improvement
> > of precursor structures, the function is not complex, regardless of
> > what it looks like.
>
> Not at all. The swarming function is indeed a fairly complex
> function. Though not as complex as various motility functions like
> flagellar motility, it does require a reasonably large number of
> fairly specified amino acids working at the same time. However, the
> swarming system of function did not evolve gradually in this case just
> as penicillin resistance via increased penicillinase production is not
> dependent upon the de novo evolution of the penicillinase function.
The ancestral organism couldn't swarm using A-motility. The new
organism can. That is the de novo evolution of a *function* in my
book. It is *not* the de novo evolution of a *structure*, but then
neither is a flagellum if it evolves from a pilus. Neither is a human
brain from an australopithecene braine, or a bird wing from a
maniraptor's forelimb. These are all mere modifications of already
existing structures to realize a new function. Which is how evolution
works according to everyone except Dr. Pitman.
>
> > I have asked you several times to demonstrate the neutral gaps between
> > a flagellum and any likely evolutionary precursors more than once, and
> > each time you have answered with something like: "You're the
> > evolutionist; that's your job." But here you are presenting an
> > argument that shoulders you with precisely that burden: the claim
> > that something is too complex to evolve is meaningless unless you can
> > tell me what it is too complex to evolve *from*.
>
> Look, I have said over and over again that the supposed evolutionary
> steps between functional structures such as the TTSS and the motility
> system of a flagellum are too widely spaced to be crossed by the
> mindless forces of random mutation and natural selection. Ian
> Musgrave and many others have detailed several supposed stepping
> stones, but not even one of these steps has been demonstrated to
> actually evolve in the laboratory or elsewhere in nature in real time.
I see. And we are justified in concluding that anything that hasn't
been demonstrated to happen in 10 years will never happen even in a
million years on the basis of...? Oh, yes, that's right, your cow
jumping over the house analogy. Well, that's been dealt with. The
analogy is flawed, and the inference is fallacious.
> Certainly if the density of these beneficial stepping stones were
> really as abundant as you evolutionists claim that they are, such an
> evolutionary process would be a piece of cake, quickly realized in
> just a few years. The fact is that these supposed beneficial stepping
> stone functions are actually universes apart in sequence space,
> separated by vast oceans of non-beneficial possibilities.
Please demonstrate this "fact". How many neutral gaps? A 95%
confidence interval will do fine.
Von Smith
<snip>
A change in the size and connection to other cells is indeed a
modification of a structure. And one with functional significance.
To say that there was no structural modification at all would not be
representing evidence accurately at all.
>
> > Dr. Pitman: There is no complexity here. The function only required
> > a slight, gradual modification of pre-existing structures.
>
> Correction: The function in question required no structural
> modification whatsoever - only increased production of what was
> already there.
In other words: a modification of a pre-existing structure.
>
> > t.o. regular: That's what I just said! Just about any function you
> > name, to include bacterial motility, can and has been observed to
> > evolve that way.
>
> Not at all. Bacterial motility has not been observed to evolve in
> that way or any other way - ever. Take a historically non-motile
> colony of bacteria and then get them to evolve any form of motility.
> You can't do it can you? Why not? If it is so easy and if selection
> pressures are high enough, such a motility function should evolve in
> short order. Or, evolve any other type of function of equal or
> greater complexity (minimum specified amino acid requirement of a few
> thousand) where the actual required structure is new!
The M. xanthus mutants went from a colony that could not swarm to one
that could. Hence, it evolved motility. You can whine all you want
about the fact that I am somehow cheating by citing an example where
the function didn't have to evolve completely from scratch. Evolution
usually "cheats" in precisely this manner, which is precisely the
point people keep trying to make to you again and again. To the
person observing it, it will seldom look like the sort of dramatic
changes you are looking for: those will only become apparent over
much longer time frames.
Look at the proposed evolutionary scenario for the flagellum again and
tell me: at what point does a "new structure" appear by your
standards? Certainly not the appearance and lengthening of the
flagellin filament, as this is just producing more of a protein that
was already being secreted, in a fashion very similar to the biofilm
formation you dismissed as not even being a modification of structure.
Once you have a pilus, what new structures must be made to evolve a
flagellum?
>
> > Dr. Pitman: That is not sufficient. Observing the evolution of a
> > motility function that requires only a slight, gradual modification of
> > pre-existing structures does not demonstrate that motility functions
> > can evolve via slight, gradual modifications of pre-existing
> > structures. In order to demonstrate that it is possible for motility
> > to evolve via slight, gradual modification of pre-existing structures,
> > you must be able to show that motility functions can evolve via a
> > dozen completely novel proteins appearing quickly together.
>
> You must at least modify pre-exiting structures to get novel types of
> functions, not just make more of what you already have. Where is the
> structural modification here?
Making something larger (many times larger, in fact) and connecting it
to other cells isn't a structural modification? I suppose a bat's
wing didn't really evolve, then, as all it is is making more of the
finger bones, skin, and muscle that the organism already had. I
suppose, too, that the human cranium isn't a structural modification,
as it is just more of what apes already had. And if connections among
cells doesn't count as structural modification, then I suppose the
human brain isn't really a structural modification of an ape's brain,
either, just more cerebral cortex (which apes already have) with cells
differently and more extensively connected.
Only if evolution proceeds the way you suggest, which it doesn't. And
all this crap about thousands of amino acids working together in a
fairly specified manner, etc. means nothing. Unless you can plug this
rhetoric into an actual discussion of how the proteins in question
work, what their active sites are, what the key structural motifs are,
which residues must be conserved to retain function, what domains they
are divided into, etc. you are talking out your ass. As little as I
profess to know about protein chemistry, I know this much.
As for your incredulity about how mindless evolutionary processes,
etc., it is not an argument. Remember: you are not simply expressing
skepticism about evolution, you have committed yourself to arguing
that it is *impossible*, which carries with it a much stronger burden
of proof than you have been able to support. You haven't even really
shown that you understand what the evolution of complex functions
actually entails or requires, let alone demonstrated that it is
impossible.
The conclusion that life has evolved is an inference to best
explanation of the available evidence of what has actually happened,
and evidence that something has actually happened is a fortiori
evidence that it is possible. You have claimed that your designer
explains the same evidence, but it really doesn't, except in the
totally vacuous sense that a sufficiently powerful and inscrutable
leprechaun can create any appearance.
>
> In fact, if you were correct in your ideas about the relatively high
> density of beneficial functions in sequence space, evolution at very
> high levels of complexity should be popping into existence all around
> you every day.
No, that doesn't follow, sorry.
> The fact that this is not happening should give you
> pause to consider that perhaps the density of beneficial functions in
> sequence space does not behave quite like you have imagined.
There is no such thing as a "beneficial function" simpliciter. What
is beneficial for one organism can be deadly for another (e.g.,
submersion in stagnant water), or even for the same organism in a
different context (e.g., at a different point in its life cycle). As
long as a peptide is biochemically active, there is at least some
chance that it will be beneficial. The rest depends on context and
history.
> Perhaps
> each additional minimum amino acid requirement really does increase
> the size of sequence space by 20-fold while the number of beneficial
> sequences increases by only a fraction of this amount.
But because evolution builds protein sequences modularly from
sequences that already work, rather than blindly searching sequences
of arbitrary length, this really doesn't matter.
Troy Truchon
> The M. xanthus mutants went from a colony that could not swarm to one
> that could. Hence, it evolved motility. You can whine all you want
> about the fact that I am somehow cheating by citing an example where
> the function didn't have to evolve completely from scratch. Evolution
> usually "cheats" in precisely this manner, which is precisely the
> point people keep trying to make to you again and again. To the
> person observing it, it will seldom look like the sort of dramatic
> changes you are looking for: those will only become apparent over
> much longer time frames.
Or to simplify, "Evolutionary theory does not propose that things magically
appear, its part of your ill-defined hypothesis, read your literature."
The fact is that if a monkey ever did give birth to a fully formed human
that actually might present some evidence for something like ID, although i
wouldn't hold your breath.
--
Welcome to California, the earth move for you too?
-troy