The Evolution of Boojumase

[rossum]

It is some time since I have posted this, so I thought it was time for
another outing. So, once again, we have ...

rossum



The Evolution of Boojmase
=========================

Creationists often talk about the impossible odds of evolution
producing a working protein. When they, rarely, show their
calculatins, they always have the same error in them. Evolutin is a
process including both random mutation and natural selection. The
creationist's models invarialy omit the natural selection part, and
only include the random mutation bit, what they often describe as a
"tornado in a junkyard", following Hoyle. That model is incorrect.
Using a correct model gives very different results.

Their naive "protein probability" calculation uses a very crude model
of evolution, basically it assumes the evolution of the whole protein
in a single large step. This gives a chance of one in 20^100 for a
protein with 100 amino acids. 20^100 comes to 1.27 x 10^131 so the
chance of the protein appearing in one step is 1 in 1.27 x 10^131.
The usual calculation is then to halve this number and say that in a
species that reproduces annually the protein is only likely to appear
after 6.35 x 10^130 years at 50% probability, which is far longer than
the age of the earth.

The model implied in the naive calculation is not the model used by
evolution. Under evolution small changes arise randomly, and may be
beneficial or deleterious. Once arisen they are selected very
non-randomly with the deleterious changes disappearing and the
beneficial changes spreading through the population. This has the
effect of spreading out the development of the protein so that instead
of one very large and unlikely jump there are a lot of smaller jumps
which are individually more likely. It also allows for the ratcheting
effect of natural selection, whereby any deleterious mutations tend to
disappear from the population and any beneficial mutations tend to
spread through the population and still be around when the next
mutation comes along.

Using a better model gives a time to evolve a protein with 100 amino
acids of just over two million years. The evolutionary model given
below for the evolution of boojumase is more complex than the naive
model, so it will need more complex calculations.


1 The Scenario
==============

Momerathius vulgaris, the common Mome Rath, is a sessile marine filter
feeder. It lives for one year, reproduces and dies. Its normal food
is a protist, Snarkius snarkius, the snark. However a proportion of
snarks are actually boojums, Snarkius boojum. Boojums have a
different cell wall, so are indigestible, most passing through the
Mome Rath's gut undigested. If there are too many boojums in the Mome
Rath's diet then it will softly and suddenly vanish away.

Mome Raths have an enzyme to digest the cell walls of snarks:
snarkase. The gene for snarkase is duplicated in the Mome Rath's
genome. This means that the second copy of the snarkase gene, called
"snk2", is available to evolve into a new gene to code for boojumase,
which will allow the Mome Rath to digest boojums as well. Any Mome
Rath possessing even a partially effective boojumase will have an
advantage in that it will have more food available and will have a
reduced chance of softly and suddenly vanishing away.

I will calculate the likely time to evolve a gene for boojumase from
the second copy of the snarkase gene. The answer turns out to be
2,096,000 years, a long time but certainly not impossible. This
figure is confirmed to within 2% by computer modelling.


2 Assumptions
=============

1 The population of Mome Raths is stable at about 10 million
individuals. Predation limits the population so the effectiveness of
boojumase is not a factor.

2 Both snarkase and boojumase contain 100 amino acids.

3 In a boojumase there is only one effective amino acid allowed in
each of the 100 positions, any of the nineteen other amino acids is
ineffective in that position.

4 Snarkase and boojumase are very different so their initial match is
just 5% (1 in 20). This is the level of matching expected from any
two random series of 100 amino acids.

5 I will only deal with mutations that have a real effect on the snk2
gene, called "significant mutations". This effect may be good or bad,
but there must be a real effect. Neutral mutations are ignored,
including replacing one ineffective amino acid with a different, but
still ineffective, amino acid in a given position. Mutations
affecting other genes are also ignored.

6 Every four thousand years there is a significant mutation in the
snk2 gene of one Mome Rath in the population of ten million. The
resulting change is random and may be good or bad. In each of the 100
positions replacing an ineffective amino acid with an effective amino
acid is good. Replacing an effective amino acid with an ineffective
one is bad. For the purposes of this model any beneficial mutation
that appears and then gets eaten by a predator or otherwise fails to
reproduce is ignored.

7 The effectiveness of a boojumase at digesting boojums is equal to
the percentage of effective amino acids it contains. Thus since
normal snarkase has 5% effective amino acids it can digest 5% of
boojums during their passage through the mome rath's gut. A boojumase
with 20 effective amino acids would be 20% effective and would digest
20% of boojums and so forth.

8 Each 1% of increased effectiveness of boojumase gives a Mome Rath a
1% advantage in reproduction. Similarly a 1% decrease in
effectiveness will give a 1% decrease in the effectiveness of
reproduction. This includes the effect of the changed probability of
the Mome Rath softly and suddenly vanishing away.


3 Preliminary Calculations
==========================

3.1 Beneficial Mutations
------------------------

First I will look at the spread of beneficial mutations. On average
each Mome Rath will reproduce exactly one Mome Rath that survives to
maturity and reproduces - this keeps the population stable. A Mome
Rath with a single mutation for a better boojumase will reproduce 1.01
Mome Raths (1% better than average), while one with a mutation for a
worse boojumase will only reproduce 0.99 Mome Raths (1% worse than
average). Better and worse will be relative, since the population
will stay at 10 million Mome Raths as the improved boojumase evolves.

How long will it take a single beneficial mutation to spread through
the population?

Initially the number of Mome Raths with the beneficial mutation will
increase from the original single individual as the powers of 1.01.
After 1551 years they will form half of the population, about 5
million (1.01 ^ 1551 = 5,040,234). From this point they will be the
normal population while those without the mutation will be the
minority. The minority will decrease as the powers of 0.99. After a
further 1604 years those without the mutation will be extinct, less
than half an individual in the population: 5,000,000 x (0.99 ^ 1604) =
0.4986802).

This means that it will take about 1551 + 1604 = 3155 years to replace
a population with the old snk2 gene with a new population with the new
improved snk2 gene.

3.2 Deleterious Mutations
-------------------------

If a deleterious mutation occurs then it will decrease as powers of
0.99 since it will be 1% less efficient at reproducing. From an
initial population of one single individual it will fall below 0.5 in
69 years (1 x 0.99 ^ 69 = 0.499837). The deleterious mutation will be
eliminated by natural selection in less than 100 years.

3.3 Mutation Rates
------------------

I have assumed one significant mutation every four thousand years. Is
this a reasonable assumption? The mutation rate for mammals is about
3 x 10 ^ -8 mutations per base pair per generation. Applying this
rate to M. vulgaris we would expect about 3 x 10 ^ -6 mutations in the
100 base pair snk2 gene per individual. With 10 million individuals
in the population this is 3 x 10 ^ -6 x 10 ^ 7 = 30 mutations in the
snk2 gene over the whole population in each generation. Naturally,
many of these mutations will be neutral. Over 4000 years there will
be 4000 x 30 = 120000 mutations in the snk2 gene over the whole
population.

From this it would appear that the rate of one significant mutation
every 4000 years is probably an underestimate. Nevertheless I will
retain it in order to avoid overlapping mutations and so simplify the
calculations. This means that the time calculated will be longer than
it would be if a more realistic rate for significant mutations was
used.


4 Mutate!
=========

4.1 The First Three Mutations
-----------------------------

Every four thousand years there is a significant mutation in the snk2
gene of one individual Mome Rath. Since the initial snarkase has 5%
effective amino acids, the first significant mutation will have a 95%
chance of being beneficial; switching an ineffective amino acid to an
effective one. It will have a 5% chance of being deleterious;
switching an effective amino acid to an ineffective one.

Deleterious mutations will disappear in 69 years, so they will be gone
before the next significant mutation in 4,000 years. Beneficial
mutations spread through the entire population in 3,155 years, so the
entire population will have the improved boojumase before the next
significant mutation. This means that significant mutations will not
overlap.

Deleterious mutation will disappear and so will not change the
probabilities for the next mutation in 4,000 years; the boojumase will
be unchanged. Beneficial mutations will be preserved and so will
increase the probability of a subsequent deleterious mutation by 1%
and reduce the probability of a beneficial mutation in 4,000 years
time. Boojumase will now be 1% more effective than before.

Drawing up the first three mutations in tables (best in a monospaced
font like Courier):

Mutation 1 (Year 4000) Mutation 2 (8000) Mutation 3 (12000)

D = 5% DD = 5% x 5% DDD = 5% x 5% x 5%
B = 95% DB = 5% x 95% DDB = 5% x 5% x 95%
BD = 95% x 6% DBD = 5% x 95% x 6%
BB = 95% x 94% DBB = 5% x 95% x 94%
BDD = 95% x 6% x 6%
BDB = 95% x 6% x 94%
BBD = 95% x 94% x 7%
BBB = 95% x 94% x 93%

Here D is a deleterious mutation and B is a beneficial mutation.
Taking an example, BBD in the seventh row of the third table, there is
a beneficial mutation followed by another beneficial mutation followed
by a deleterious mutation. The first beneficial mutation has a
probability of 95%, the second beneficial mutation only has a
probability of 94% since the boojumase now has 6 effective amino acids
and 94 ineffective ones as it was improved by the first beneficial
mutation. The probability of the final deleterious mutation is 7%
since there are seven effective amino acids in the boojumase after two
beneficial mutations.

4.2 Average Expected Effectiveness
----------------------------------

Tracing this through many mutations will result in huge tables: 2 ^
100 rows after 100 mutations. In order to proceed I am going to
simplify the calculation by working out a single "Average Expected
Effectiveness" (AEE) for the effectiveness of the boojumase. Doing
some more calculations on the table for the third mutation gives:

Prob. Effect. P x E
DDD 0.0125% 5% 0.000625%
DDB 0.2375% 6% 0.014250%
DBD 0.2850% 6% 0.017100%
DBB 4.4650% 7% 0.312550%
BDD 0.3420% 6% 0.020520%
BDB 5.3580% 7% 0.375060%
BBD 6.2510% 7% 0.437570%
BBB 83.0490% 8% 6.643920%
-------- ---------
100.0000% 7.821595% = AEE

Here the "Prob." column is the probability of that particular outcome
for the three mutations; for example the probability of DBD is 5% x
95% x 6% = 0.2850%. The sum of the probabilities is 100% as a check
on the calculation. The "Effect." column is the effectiveness of the
boojumase after the mutations; start at 5% and add 1% for each B, so
DDD is still at 5% effectiveness while BBB is at the maximum possible
8% effectiveness after three beneficial mutations. The "P x E" column
is the previous two columns multiplied together and adjusted to a
percentage. Each entry is the proportion of the effectiveness that
this row contributes to the overall expected effectiveness of the
boojumase. The sum of this column is the "average expected
effectiveness" (AEE) that I wish to calculate: 7.82 to two decimal
places.

4.3 The Fourth Mutation
-----------------------

Coming into the fourth mutation the average expected effectiveness
(AEE) is 7.82. This gives a 7.82% chance of a deleterious mutation
and a (100.00 - 7.82) = 92.18% chance of a beneficial mutation. The
table looks like:

Mutation 4 (16000) Initial AEE = 7.82%

Prob. Effect. P x E
D 7.82% 7.82% 0.61%
B 92.18% 8.82% 8.13%
------- -----
100.00% 8.74% = new AEE

The probability of a deleterious mutation, D, is the AEE, 7.82. The
probability of a beneficial mutation, B, is (100 - AEE), 92.18%. The
effectiveness of the boojumase after a deleterious mutation is
unchanged, the AEE, 7.82%. The effectiveness of the boojumase after a
beneficial mutation is increased by 1%, (AEE + 1), 8.82%. The P x E
column has AEE x AEE / 100 in the D row and (100 - AEE) x (AEE + 1) /
100 in the B row. In each case the "/ 100" is to get the P x E column
back into a percentage. The new AEE is the sum of these two values:
(AEE x AEE / 100) + ((100 - AEE) x (AEE + 1) / 100). This is the new
value of the AEE to go forward to the next mutation.

4.4 The Fifth Mutation
----------------------

From the discussion of the fourth mutation there is a formula for
calculating the AEE after fifth mutation. The formula is:

New AEE = (AEE x AEE / 100) + ((100 - AEE) x (AEE + 1) / 100)

This can be simplified to:

New AEE = ((99 x AEE) + 100) / 100

Putting the AEE of 8.74 coming into the fifth mutation into the
formula gives 9.65 to two decimal places for the AEE after the fifth
mutation.

4.5 And so on...
----------------

The simplified formula from section 4.4 can be used to step from
mutation to mutation. The calculation is best shown in a table. Rows
are missed out purely for reasons of space. It is simple to set up
the whole thing on a spreadsheet.

Year Mutation AEE after
12000 3 7.82%
16000 4 8.74%
20000 5 9.65%
40000 10 14.08%
80000 20 22.30%
100000 25 26.11%
200000 50 42.52%
500000 125 72.95%
800000 200 87.27%
1000000 250 92.30%
2000000 500 99.38%
2096000 524 99.51%

This shows that after a million years of evolution and 250 significant
mutations M. vulgaris has a snk2 gene that codes for a boojumase that
is on average 92% effective. 92 of the hundred amino acids in the
boojumase are effective, on average only eight are ineffective. After
two million years 99 of the hundred amino acids are effective with an
average of one ineffective amino acid.

The table also shows that as the boojumase becomes more effective,
random mutations are more likely to be deleterious, so it takes longer
between beneficial mutations.


5 Result
========

After an average of 2,096,000 years and 524 significant mutations the
Mome Raths have evolved the most effective boojumase possible as there
is less than half an amino acid that is ineffective on average. The
Mome Raths have adapted to an environment containing boojums and will
not softly and suddenly vanish away.

This average figure of 2,096,000 years to evolve a protein with 100
amino acids compares with the 6.35 x 10^130 years calculated from the
less realistic naive model that failed to account for the non-random
element of natural selection.


6 Computer Modelling
====================

Putting this model into a computer program and running it through to
the evolution of a 99.5% effective boojumase a million times gave the
results:

Mean Mutations Std Deviation
513.74 125.97

Running the program three more times, each with a million repetitions
gave:

Mean Mutations Std Deviation
513.65 125.89
513.71 125.79
513.70 125.86

This seems to indicate that the calculations above are a little
pessimistic, and the average should be 514 mutations, taking 2,056,000
years instead of 524 mutations taking 2,096,000 years. The error is
less than two percent. No doubt a better mathematician or
statistician than me could explain the discrepancy.


7 The Boojumase Model
=====================

This is a simple model, deliberately so in order to simplify the
calculations. However it is more complex and closer to the real
situation than the model implied by the naive probability calculation.
The naive model covers the random nature of mutations but it does not
include either the highly non-random process of natural selection or
the ratcheting effect of small changes over the generations in a
population and so gives a misleading result. The boojumase model by
including random mutations, the non-random element of natural
selection and the ratchet effect gives a less misleading result.

The boojumase model is intended as a learning aid. For that reason it
is simplified to remove all calculus and more advanced mathematics.
It is intended for an interested lay audience, not for publication in
Science or Nature.

I have deliberately made life difficult for the model by starting with
the 5% random match between snarkase and boojumase, by allowing only
one amino acid to be effective at each position and by picking a long
interval between significant mutations. This is to avoid criticism
that the model is biased in favour of a short time to evolve the
protein; if anything the model is biased towards a long time to evolve
the protein.

The model is by no means perfect. Possible improvements to it are:

- To improve the calculation of the time taken to spread a beneficial
mutation through the whole population. I tried this myself and got a
figure of 3293 years; not different enough to warrant the extra
complexity and with no effect on the overall result as it is still
less than 4,000 years.

- To take into account sexual reproduction in the spreading of
beneficial mutations.

- Run the exact calculation of tables for more than three mutations
before switching to the AEE.

- Explain the transition from the exact tables to the AEE better than
I have in section 4.2.

- Allow mutations to overlap so a second significant mutation might
occur before the previous mutation has spread through the whole
population. This would allow a more realistic rate of significant
mutations.

- Look at mutation rates in real life and make a better assumption
for the interval between significant mutations. I picked 4000 years
purely to avoid complications with overlapping mutations.

Feel free to take up this model, clean it up and make it a better
reflection of reality. If you do so please bear in mind its purpose
and do not complicate it too much; remember the target audience.


8 Bibliography
==============

Lewis Carroll: Jabberwocky

Lewis Carroll: The Hunting of the Snark

[Mitchell Coffey]

On Nov 18, 10:57 am, rossum <rossu...@coldmail.com> wrote:
> It is some time since I have posted this, so I thought it was time for
> another outing.  So, once again, we have ...
>
> rossum
>
> The Evolution of Boojmase
> =========================

[snup]

Allow me to be the first to say: brillig!

Mitchell Coffey

[r norman]

On Fri, 18 Nov 2011 15:57:56 +0000, rossum <ross...@coldmail.com>
wrote:

Here is a photo of the devastating effect boojumase has on what would
otherwise be a normal tree
http://www.treepicturesonline.com/boojum-tree-picture4.jpg

[John S. Wilkins]

I have to say, it left me in its wabe.

--
John S. Wilkins, Associate, Philosophy, University of Sydney
http://evolvingthoughts.net
But al be that he was a philosophre,
Yet hadde he but litel gold in cofre

[Paul J Gans]

rossum <ross...@coldmail.com> wrote:
>It is some time since I have posted this, so I thought it was time for
>another outing. So, once again, we have ...

>rossum



>The Evolution of Boojmase
>=========================

[...]

Excellent! Amazingly good!

--
--- Paul J. Gans

[r norman]

On Sat, 19 Nov 2011 10:57:57 +1100, jo...@wilkins.id.au (John S.
Wilkins) wrote:

>Mitchell Coffey <mitchel...@gmail.com> wrote:
>
>> On Nov 18, 10:57 am, rossum <rossu...@coldmail.com> wrote:
>> > It is some time since I have posted this, so I thought it was time for
>> > another outing. So, once again, we have ...
>> >
>> > rossum
>> >
>> > The Evolution of Boojmase
>> > =========================
>> [snup]
>>
>> Allow me to be the first to say: brillig!
>>
>I have to say, it left me in its wabe.

Would it be ever so pedantic to remark that Jabberwocky is contained
in a work quite separate from the Hunting of the Snark?

[alextangent]

On Nov 19, 12:23 am, r norman <r_s_nor...@comcast.net> wrote:
> On Sat, 19 Nov 2011 10:57:57 +1100, j...@wilkins.id.au (John S.
>
>
>
>
>
>
>
>
>
> Wilkins) wrote:

> >Mitchell Coffey <mitchell.cof...@gmail.com> wrote:
>
> >> On Nov 18, 10:57 am, rossum <rossu...@coldmail.com> wrote:
> >> > It is some time since I have posted this, so I thought it was time for
> >> > another outing.  So, once again, we have ...
>
> >> > rossum
>
> >> > The Evolution of Boojmase
> >> > =========================
> >> [snup]
>
> >> Allow me to be the first to say: brillig!
>
> >I have to say, it left me in its wabe.
>
> Would it be ever so pedantic to remark that Jabberwocky is contained
> in a work quite separate from the Hunting of the Snark?

'You should learn not to make personal remarks,' Alice said with some
severity; 'it's very rude.'

[Mitchell Coffey]

On 11/18/2011 6:57 PM, John S. Wilkins wrote:
> Mitchell Coffey<mitchel...@gmail.com> wrote:
>
>> On Nov 18, 10:57 am, rossum<rossu...@coldmail.com> wrote:
>>> It is some time since I have posted this, so I thought it was time for
>>> another outing. So, once again, we have ...
>>>
>>> rossum
>>>
>>> The Evolution of Boojmase
>>> =========================
>> [snup]
>>
>> Allow me to be the first to say: brillig!
>>
> I have to say, it left me in its wabe.

Quite a flight of mimsy.

Mitchell

[John S. Wilkins]

Holy scripture comes in many books:

Alice's Adventures in Wonderland (1865)
Facts
Rhyme? And Reason? (also published as Phantasmagoria)
Pillow Problems
Sylvie and Bruno
Sylvie and Bruno Concluded
The Hunting of the Snark (1876)
Three Sunsets and Other Poems
Through the Looking-Glass, and What Alice Found There (includes
"Jabberwocky" and "The Walrus and the Carpenter") (1871)
What the Tortoise Said to Achilles

Also:
Winnie-the-Pooh (1926) (illustrated by E. H. Shepard)
The House at Pooh Corner (1928) (illustrated by E. H. Shepard)
When We Were Very Young
Now We Are Six

Trial by Jury (1875)
The Sorcerer (1877)
H.M.S. Pinafore, or The Lass that Loved a Sailor (1878)
The Pirates of Penzance, or The Slave of Duty (1880)
Patience, or Bunthorne's Bride (1881)
Iolanthe, or The Peer and the Peri (1882)
Princess Ida, or Castle Adamant (1884)
The Mikado, or The Town of Titipu (1885)
Ruddigore, or The Witch's Curse (1887)
The Yeomen of the Guard, or The Merryman and his Maid (1888)
The Gondoliers, or The King of Barataria (1889)
Utopia Limited, or The Flowers of Progress (1893)
The Grand Duke, or The Statutory Duel (1896)

There are a number of minor scriptures, such as those by the Prophets
Kafka (PBUH) and Asimov (PBUH).

Defer!

[Walter Bushell]

In article
<4dd13a23-bcd9-4d0c...@r28g2000yqj.googlegroups.com>,

Stuff like this makes me want to outgrabe.

--
It is the nature of the human species to reject what is true but unpleasant
and to embrace what is obviously false but comforting. -- H. L. Mencken

[Free Lunch]

On Sat, 19 Nov 2011 12:41:23 +1100, jo...@wilkins.id.au (John S. Wilkins)
wrote in talk.origins:

A very fine collection of scripture.

[r norman]

On Sat, 19 Nov 2011 12:41:23 +1100, jo...@wilkins.id.au (John S.

Sorry but my posts was merely corroborative detail, intended to give
artistic verisimilitude to an otherwise bald and unconvincing
narrative.

As to the canon: "Very young" and "Now we are.." were childhood works,
completely assimilated but only half memorized in grade school. The
Penguin "Complete works of Lewis Carroll" was finished in high school,
cover to cover. I never saw "Grand Duke" but did see "Cox and Box"
(only semi-canonical) having been a long time FUMGASS (friend of the
UofM G&S Society).

You forgot Gamow -- "1,2,3 Infinity" and the incomparable "Mr.
Tompkins" books.

[Mitchell Coffey]

[cough]Pratchett[cough]

Mitchell

[John S. Wilkins]

Different religion. These are the Holy Writ of Philosophical Examples. I
should have put Walt Whitman in there as well.

[Michael Siemon]

In article <1kazni5.1sj0w961qpm39lN%jo...@wilkins.id.au>,

Out of the Kraken, endlessly Rocking? I'm boggled at your introducing
Whitman here (despite my fondness for him). Somehow, "Crossing Brooklyn
Ferry" and "When Lilacs Last in the Courtyard Bloomed" don't seem to
fit with these others. I suppose the "Do I contradict myself" line might
be commentary on our discussions, but -- huh? --

[John S. Wilkins]

Song of Myself has dozens of lines used by philosophers.

E.g.,

Has any one supposed it lucky to be born?
I hasten to inform him or her it is just as lucky to die, and I know it.

I pass death with the dying and birth with the new-wash'd babe, and
am not contain'd between my hat and boots

or

I accept Reality and dare not question it,
Materialism first and last imbuing.

Hurrah for positive science! long live exact demonstration!
Fetch stonecrop mixt with cedar and branches of lilac,
This is the lexicographer, this the chemist, this made a grammar of
the old cartouches,
These mariners put the ship through dangerous unknown seas.
This is the geologist, this works with the scalper, and this is a
mathematician.

Gentlemen, to you the first honors always!
Your facts are useful, and yet they are not my dwelling,
I but enter by them to an area of my dwelling.

[Michael Siemon]

In article <1kazov5.1gy4g7h5xqf9hN%jo...@wilkins.id.au>,

OK. It certainly is a grab bag of stuff.. :-) I hadn't encountered
that kind of usage (but my philosophy background is of a rather odd
"Chicago School" [no, _not_ the business/economic kind! :-)]) that
stemmed from R. M. Hutchins' notions of what a university was about...
That and the Committee on Social Thought.

[Mitchell Coffey]

[cough]Twain[cough]

Mitchell

[Mitchell Coffey]

There's poems for philosophy, then there're poems for politics:

Democracy

The flag goes with the foul landscape, and our jargon muffles the drum.

In the great centers we'll nurture the most cynical prostitution. We'll
massacre logical revolts.

In spicy and drenched lands!-- at the service of the most monstrous
exploitations, industrial or military.

Farewell here, no matter where. Conscripts of good will, ours will be a
ferocious philosophy; ignorant as to science, rabid for comfort; and let
the rest of the world croak. This is the real advance. Marching orders,
let's go!

- Arthur Rimbaud, Illuminations


Mitchell

[Steven L.]

"Mitchell Coffey" <mitchel...@gmail.com> wrote in message
news:4dc435b0-03a2-4b2e...@q16g2000yqn.googlegroups.com:

Yep.
My toves are slithy.




-- Steven L.

[Karel]

Interesting, no, very interesting as this is, aren't you
ignoring the possibility that the development might get
stuck at some fitness optimum where no single beneficial
mutation is possible?

Regards,

Karel

[Free Lunch]

On Fri, 18 Nov 2011 22:31:13 -0800, Michael Siemon <mlsi...@sonic.net>
wrote in talk.origins:

Seems like a great program from what I see about it, but the name is a
bit unsettling.

[jillery]

Now now, no need to talk dirty :)

[rmj]

>
> 7 The effectiveness of a boojumase at digesting boojums is equal to
> the percentage of effective amino acids it contains. Thus since
> normal snarkase has 5% effective amino acids it can digest 5% of
> boojums during their passage through the mome rath's gut. A boojumase
> with 20 effective amino acids would be 20% effective and would digest
> 20% of boojums and so forth.
>
> 8 Each 1% of increased effectiveness of boojumase gives a Mome Rath a
> 1% advantage in reproduction. Similarly a 1% decrease in
> effectiveness will give a 1% decrease in the effectiveness of
> reproduction. This includes the effect of the changed probability of
> the Mome Rath softly and suddenly vanishing away.
>

These are erroroneous, contrary to reality, assumptions. Indeed, this
error is fundamental to much of the theory concerning evolutionary
mechanisms. There is no way that one takes a protein for one metabolic
process, makes a single amino acid change and gets a 1% increase in a
different metabolic process. To get an active site on a protein there is
little leeway on changes to that protein; besides requiring effective
amino acids at the active site, most other amino acids must contribute
to a correct folding of the protein to get the active site.

This notion of gradualism is a fantasy.

[rossum]

On Sat, 19 Nov 2011 12:41:23 +1100, jo...@wilkins.id.au (John S.

There is only one true and Holy Scripture, completely free from error:

The Bellman's Map
-----------------

He had bought a large map representing the sea,
Without the least vestige of land:
And the crew were much pleased when they found it to be
A map they could all understand.

"What’s the good of Mercator’s North Poles and Equators,
Tropics, Zones, and Meridian Lines?"
So the Bellman would cry: and the crew would reply
"They are merely conventional signs!

"Other maps are such shapes, with their islands and capes!
But we’ve got our brave Bellman to thank"
(So the crew would protest) "that he’s bought us the best -
A perfect and absolute blank!"

rosuum

[rossum]

On Sat, 19 Nov 2011 07:36:49 -0800 (PST), Karel
<GCPAXS...@spammotel.com> wrote:

>Interesting, no, very interesting as this is, aren't you
>ignoring the possibility that the development might get
>stuck at some fitness optimum where no single beneficial
>mutation is possible?
>
>Regards,
>
>Karel

This is a simple teaching exercise, intended to illustrate the
differentce between the typical creationist "tornado in a junkyard"
scenario and a simple calculation that includes the effects of natural
selection.

In real life, you are right of course. Evolution does get suck at
local optima; it even remains on those local optima when they are no
longer optimal. Think of recurrent laryngeal nerve which was
optimally routed when we were fish, but now has a sub-optimal route
because it has kept to the one optimal route.

rossum

[rossum]

On Sat, 19 Nov 2011 12:22:43 -0800, rmj <glennaRe...@jps.net>
wrote:

I said that this was an illustration of the incorrectness of "tornado
in a junkyard" models that omit natural selection. Even YECs accept
the validity of micro-evolution, and almost all of micro-evolution is
gradual. Minor tweaks spreading through a population.

Have you looked at Yockey's calculation of the number of possible
Cytochrome-C's? 2.3 x 10^93 possible ways to make a working
Cytochrome-C. There is plenty of gradualism allowed ithin that number
of options.

rossum

[Karel]

On 20 nov, 14:09, rossum <rossu...@coldmail.com> wrote:
> On Sat, 19 Nov 2011 07:36:49 -0800 (PST), Karel
>

Thank you for the explanation. And as a teaching exercise,
it certainly worked for me. Thank you!

Regards,

Karel

[Paul J Gans]

I certainly do!

[Paul J Gans]

Actually, if one wants to add stuff, I'd add all (or almost all))
of GBShaw's prefaces to his plays.

[Mark Isaak]

Of that, there is no possible doubt,
no probable, possible shadow of doubt,
no possible doubt whatever.

--
Mark Isaak eciton (at) curioustaxonomy (dot) net
"It is certain, from experience, that the smallest grain of natural
honesty and benevolence has more effect on men's conduct, than the most
pompous views suggested by theological theories and systems." - D. Hume

[rmj]

I haven't looked at Yockey. His calculation is undoubtedly based on a
number of assumptions. Still, if one accepts is number as fairly
accurate, it is relatively miniscule. For a 100 amino acid protein (like
cytochrome c), there are 20^100 possible amino acid sequences. Now my
math skills are not solid, but that is of the order of 10^130. So
compare 10^93 to 10^130.

There is another problem with your model. Just because four or five
amino acid changes may increase effectiveness, does not mean that there
is another amino acid change that can increase effectiveness. How does
your model account for dead ends, lines that can get to, say, 20%
efficiency, but where there is no route to greater efficiency without
undoing what changes had occurred?

[rossum]

On Sun, 20 Nov 2011 22:41:05 -0800, rmj <glennaRe...@jps.net>
wrote:

Yockey's calculations show that the naive 10^130 figure is wrong by a
factor of 10^93. Of course the 10^130 figure includes all possible
functional proteins of that length, a figure considerably higher than
10^93. If a protein is useful then natural selection will tend to
increase its frequency in the population. It may not be useful as a
Cytochrome-C, but it may be useful as something else.

>
>There is another problem with your model. Just because four or five
>amino acid changes may increase effectiveness, does not mean that there
>is another amino acid change that can increase effectiveness. How does
>your model account for dead ends, lines that can get to, say, 20%
>efficiency, but where there is no route to greater efficiency without
>undoing what changes had occurred?

My model was deliberately kept simple. It only needed to be simple
because the creationist "tornado in a junkyard", your 10^130 is itself
grossly oversimplified and not a good model of evolution.

By all means add the possibility of getting 'caught' to my model. You
will need to explain the changes you are making and to do the forward
calculations yourself.

As I said, this is a simplified teachimg model, not an accurate model
of evolution.

rossum

[rmj]

Assertions and assumptions.

>
>>
>> There is another problem with your model. Just because four or five
>> amino acid changes may increase effectiveness, does not mean that there
>> is another amino acid change that can increase effectiveness. How does
>> your model account for dead ends, lines that can get to, say, 20%
>> efficiency, but where there is no route to greater efficiency without
>> undoing what changes had occurred?
> My model was deliberately kept simple. It only needed to be simple
> because the creationist "tornado in a junkyard", your 10^130 is itself
> grossly oversimplified and not a good model of evolution.

Then you admit you have chosen those assumptions because you have an agenda.

>
> By all means add the possibility of getting 'caught' to my model. You
> will need to explain the changes you are making and to do the forward
> calculations yourself.
>
> As I said, this is a simplified teachimg model, not an accurate model
> of evolution.

I would that teachers attempt to place honesty first.
>
> rossum
>

[rossum]

On Wed, 23 Nov 2011 07:45:29 -0800, rmj <glennaRe...@jps.net>

You are free to read Yockey's work and to point out his errors. As to
the functionality of random protiens see
http://amesteam.arc.nasa.gov/Research/proteins.html

These are not assertions and assumptions, they are considered results
from scientific experiments. Your creationist sources are lying to
you.

>>
>>>
>>> There is another problem with your model. Just because four or five
>>> amino acid changes may increase effectiveness, does not mean that there
>>> is another amino acid change that can increase effectiveness. How does
>>> your model account for dead ends, lines that can get to, say, 20%
>>> efficiency, but where there is no route to greater efficiency without
>>> undoing what changes had occurred?
>> My model was deliberately kept simple. It only needed to be simple
>> because the creationist "tornado in a junkyard", your 10^130 is itself
>> grossly oversimplified and not a good model of evolution.
>
>Then you admit you have chosen those assumptions because you have an agenda.

I am building a simple teaching model. That is what I said in my OP.
Didn't you read it?

>
>>
>> By all means add the possibility of getting 'caught' to my model. You
>> will need to explain the changes you are making and to do the forward
>> calculations yourself.
>>
>> As I said, this is a simplified teachimg model, not an accurate model
>> of evolution.
>
>I would that teachers attempt to place honesty first.

You don't start a First Grade Math class on Tensor Algebra. One of
the skills required of a teacher is to ensure that the lesson is at an
appropriate level for the pupils in the classroom. Do you have a
problem with that?

Is a Sunday School teacher who fails to teach First Graders about the
difference between 'homoousion' and 'homoiousion', and all the Greek
grammar neded to understand that differeence, being dishonest?

A full mathematical model of evolution requires calculus. My model is
intended for people who may not have a sufficient understanding of
calculus. My model is reasonably close to the real thing, and is
sufficient to show that the creationist model is grossly in error.

If you want teachers to be honest, then stop them trying to push the
creationist "tornado in a junkyard" model. That is far more dishonest
than my Boojumase model. Didn't someone say something about motes and
beams?

rossum