Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

Pseudogenes: Their Evolution and Their Functions

777 views
Skip to first unread message

Peter Nyikos

unread,
Sep 29, 2016, 11:10:02 AM9/29/16
to talk-o...@moderators.isc.org
This is a reply to a post by George Kaplan on the rapidly expanding
thread [over 800 posts now],
Subject: Re: Why cannot we see evolution happening today?

On https://groups.google.com/d/msg/talk.origins/mq1OKVRyRp4/Xu-PjO8qAAAJ on Wednesday, September 28, 2016 at 10:00:03 PM UTC-4,gkaplan wrote:
> On 9/28/2016 6:11 PM, Peter Nyikos wrote:
> > On Wednesday, September 28, 2016 at 4:00:02 PM UTC-4, gkaplan wrote:
> >> On 9/28/2016 11:26 AM, John Harshman wrote:
> >>> On 9/28/16 10:58 AM, gkaplan wrote:
> >>>> On 9/28/2016 10:24 AM, John Harshman wrote:
> >>>>> On 9/28/16 9:49 AM, gkaplan wrote:
> >>>>>

[I don't know who wrote the following:]
> >>>>>>> Why don't you calculate for us the probability that exactly 55 base
> >>>>>>> pairs would be deleted from *exactly* the same spot in the genome
> >>>>>>> among all these various primates, without common ancestry being
> >>>>>>> involved?

I'm not sure what this is supposed to refer to. My guess is that
"the same spot" refers to the same *locus* -- in this case, the
same pseudogene -- judging from what comes next.

> >>>>>> Have you considered the source could be genetic engineering and
> >>>>>> artificial selection?
> >>>>>
> >>>>> Why would anyone do genetic engineering and artificial selection in
> >>>>> order to fix a useless pseudogene mutation in a collection of unrelated
> >>>>> species?
> >>>>
> >>>> What is being fixed?
> >>>>
> >>> "Fixed", in biology, refers to a particular feature increasing in
> >>> frequency in the population until it's found in 100% of the population.
> >>> In this case, a deletion in a pseudogene is fixed in almost all primate
> >>> species.

It seems you are talking here about a deletion of a whole fragment of a
[pseudo]gene, George.

If so, which primates still have that fragment? And is it part of a
gene that codes for a protein in them?

In which sub-clade of the order *Primates* have those 55 base pairs
gone missing?

If you don't know the answers, George, someone reading this might
know them.

> >>
> >> Somehow you come to the conclusion that because a pseudogene exists in
> >> an organism that if genetic engineering was the instrument, that the
> >> goal was to create a pseudogene? Are you attempting to be facetious?

[here is where I came in:]

> > Why should he be? Don't you know that some pseudogenes, or
> > "junk DNA" as it is popularly called, have been found to have useful
> > functions, just not the function of coding for polypeptides.
> > This has been called a triumph of ID theory by some, who make
> > the claim that ID theory had predicted that a lot of "junk DNA"
> > did fulfill some useful design plan.
> >
> > Did you ever see my reply to you where I shot down something Norman
> > had convinced you of, namely that deleterious mutations are synonymous
> > with mutations that produce pseudogenes?
>
> I remember some discussion on that. I don't always try to argue every
> minor point, as it is a waste of time. I don't recall believing it the
> way you express it.

Norman seemed to think you had agreed on it. It'll take me time to
find the right place in the voluminous [see opening comment] "parent
thread."

> From what I have read, a deleterious mutation can
> also kill the organism or lower its fitness.

Yes, by the mutant allele coding for the wrong kind of protein.

But perhaps mutated pseudogenes can have the same deleterious effects.

> The way I was was using the term was specific to my argument regarding a
> completely new function being created with multiple beneficial mutations
> to do anything significant on the same locus.

In that case, why did you keep talking about 55 base pairs missing from
that pseudogene?

> In that context it does
> not impact my argument if the gene is disabled or if it kills the
> organism. That trajectory is ended.

But the question at the beginning evidently does not have anything
to do with this. A huge bunch of trajectories are very active
despite the presence of the pseudogene.

Peter Nyikos
Professor, Dept. of Mathematics -- standard disclaimer--
University of South Carolina
http://people.math.sc.edu/nyikos/

rsNorman

unread,
Sep 29, 2016, 11:35:01 AM9/29/16
to talk-o...@moderators.isc.org
Peter Nyikos <nyi...@bellsouth.net> Wrote in message:
Thank you for starting a new thread and starting "clean" with one
specific subject.

I think that your reading of what I once wrote about "pseudogenes"
is based on a misconception about what I really did write,
probably enhanced by what was probably careless wording on my
part. I never wrote that deleterious mutations are synonymous
with mutations that produce pseudogenes. I was merely trying to
interpret what many of us have had on George's use of the word
"deleterious." I was trying to indicate that a mutation in one
copy of a duplicated gene that was "deleterious" in the sense
that it destroyed that copy's ability to produce a gene product
resulted in a pseudogene. The word "deleterious" was always used
by George in this context to refer to mutations in one of the
copies and I used it that way, also.

The discussion of gene duplication did get off track with a lot of
nonsense about pseudogenes and genetic engineering.


Frankly, the whole question of gene duplication was a side issue
that George somehow seized upon thinking we "evolutionists"
claimed it was the answer to how humans and chimps could have
diverged in only 6 million years. Gene duplication can produce
genes with new functions but it is quite true that most duplicate
copies of genes do not provide any new functionality, nor any
functionality at all and end up as pseudogenes. Humans have a
estimated 20,000 pseudogenes. If only a tiny fraction of the
copies did produce some new function that would leave hundreds or
even thousands of new functions produced that way.


----Android NewsGroup Reader----
http://usenet.sinaapp.com/

John Harshman

unread,
Sep 29, 2016, 11:50:02 AM9/29/16
to talk-o...@moderators.isc.org
On 9/29/16 8:06 AM, Peter Nyikos wrote:
> This is a reply to a post by George Kaplan on the rapidly expanding
> thread [over 800 posts now],
> Subject: Re: Why cannot we see evolution happening today?
>
> On https://groups.google.com/d/msg/talk.origins/mq1OKVRyRp4/Xu-PjO8qAAAJ on Wednesday, September 28, 2016 at 10:00:03 PM UTC-4,gkaplan wrote:
>> On 9/28/2016 6:11 PM, Peter Nyikos wrote:
>>> On Wednesday, September 28, 2016 at 4:00:02 PM UTC-4, gkaplan wrote:
>>>> On 9/28/2016 11:26 AM, John Harshman wrote:
>>>>> On 9/28/16 10:58 AM, gkaplan wrote:
>>>>>> On 9/28/2016 10:24 AM, John Harshman wrote:
>>>>>>> On 9/28/16 9:49 AM, gkaplan wrote:
>>>>>>>
>
> [I don't know who wrote the following:]
>>>>>>>>> Why don't you calculate for us the probability that exactly 55 base
>>>>>>>>> pairs would be deleted from *exactly* the same spot in the genome
>>>>>>>>> among all these various primates, without common ancestry being
>>>>>>>>> involved?
>
> I'm not sure what this is supposed to refer to. My guess is that
> "the same spot" refers to the same *locus* -- in this case, the
> same pseudogene -- judging from what comes next.

Maybe you should stop responding to comments that are 7 or 8 layers
deep. That might improve your understanding. It refers to a single
deletion of a 55-base piece of the pseudogene.

>>>>>>>> Have you considered the source could be genetic engineering and
>>>>>>>> artificial selection?
>>>>>>>
>>>>>>> Why would anyone do genetic engineering and artificial selection in
>>>>>>> order to fix a useless pseudogene mutation in a collection of unrelated
>>>>>>> species?
>>>>>>
>>>>>> What is being fixed?
>>>>>>
>>>>> "Fixed", in biology, refers to a particular feature increasing in
>>>>> frequency in the population until it's found in 100% of the population.
>>>>> In this case, a deletion in a pseudogene is fixed in almost all primate
>>>>> species.
>
> It seems you are talking here about a deletion of a whole fragment of a
> [pseudo]gene, George.

That wasn't George, that was me. Once more, perhaps you shouldn't try to
go so far back in time, which might improve your understanding of what
you read.

> If so, which primates still have that fragment? And is it part of a
> gene that codes for a protein in them?

It's part of a pseudogene. If I recall, this happened in the common
ancestor of all primates in which it's a pseudogene. But I could be wrong.
It's conceivable. Any sequence might have a mutation that would produce
some deleterious effect, even a sequence that's been evolving neutrally
for millions of years. Unlikely, though. I think what George is talking
about is a mutation that would have been deleterious in a functional
protein but that's neutral in a pseudogene. He may not realize this.

>> The way I was was using the term was specific to my argument regarding a
>> completely new function being created with multiple beneficial mutations
>> to do anything significant on the same locus.
>
> In that case, why did you keep talking about 55 base pairs missing from
> that pseudogene?

I don't think he's been able to engage with the idea of a 55-base deletion.

>> In that context it does
>> not impact my argument if the gene is disabled or if it kills the
>> organism. That trajectory is ended.
>
> But the question at the beginning evidently does not have anything
> to do with this. A huge bunch of trajectories are very active
> despite the presence of the pseudogene.

Yes, he is exceeding confused, and yet he seems to think that everyone
else is confused.

gkaplan

unread,
Sep 29, 2016, 11:55:01 AM9/29/16
to talk-o...@moderators.isc.org
On 9/29/2016 8:06 AM, Peter Nyikos wrote:
> This is a reply to a post by George Kaplan on the rapidly expanding
> thread [over 800 posts now],
> Subject: Re: Why cannot we see evolution happening today?
>
> On https://groups.google.com/d/msg/talk.origins/mq1OKVRyRp4/Xu-PjO8qAAAJ on Wednesday, September 28, 2016 at 10:00:03 PM UTC-4,gkaplan wrote:
>> On 9/28/2016 6:11 PM, Peter Nyikos wrote:
>>> On Wednesday, September 28, 2016 at 4:00:02 PM UTC-4, gkaplan wrote:
>>>> On 9/28/2016 11:26 AM, John Harshman wrote:
>>>>> On 9/28/16 10:58 AM, gkaplan wrote:
>>>>>> On 9/28/2016 10:24 AM, John Harshman wrote:
>>>>>>> On 9/28/16 9:49 AM, gkaplan wrote:
>>>>>>>
>
> [I don't know who wrote the following:]
>>>>>>>>> Why don't you calculate for us the probability that exactly 55 base
>>>>>>>>> pairs would be deleted from *exactly* the same spot in the genome
>>>>>>>>> among all these various primates, without common ancestry being
>>>>>>>>> involved?
>
> I'm not sure what this is supposed to refer to. My guess is that
> "the same spot" refers to the same *locus* -- in this case, the
> same pseudogene -- judging from what comes next.

I believe this is John. The argument as I understand it, is that since
Chimps and Gorillas and Humans show the same 55 bp deletion, that this
proves common ancestry.

My response (in so many words) is that if a genetic engineer took the
DNA from one organism and added a change, that what is seen would be
consistent with this possibility. In other words, common ancestry is not
the only possibility.
That was not me and part of a different conversation.

John Harshman

unread,
Sep 29, 2016, 1:45:01 PM9/29/16
to talk-o...@moderators.isc.org
And I was indeed wrong. I looked up Vince's original claim and then
checked the literature. This is the GBA pseudogene. It's a second copy
of the functional GBA gene and is found a short distance away,
essentially a tandem repeat. Chimp and gorilla have the same pseudogene
with the same 55-base deletion. Orangutan has the second copy, but
without the deletion, and it appears to be functional. Squirrel monkey
doesn't have the second copy. The rest of the distribution in primates
is, as far as I know, unknown. But we can be fairly confident that a
duplication of GBA happened somewhere between the split between NW and
OW monkeys and the split between orangutans and other great apes, and
that the deletion (and probably the deactivation of the second copy)
happened somewhere between the split between orangutans and other great
apes and the split between gorillas and the chimp/human ancestor.

Anyway, this gives us two phylognetic markers, one for the duplication
and one for the indel. Were these markers engineered into unrelated
species, or were they just hitchhikers? I'd be interested in a developed
gene-engineering scenario that accounts for the data, plus the pattern
of genetic divergence in general.

My current opinion: I don't think one is possible unless it involves a
conscious attempt at deception on the part of the angelic engineers.

https://www.ncbi.nlm.nih.gov/pubmed/16102985

Peter Nyikos

unread,
Sep 29, 2016, 1:45:01 PM9/29/16
to talk-o...@moderators.isc.org
On Thursday, September 29, 2016 at 11:50:02 AM UTC-4, John Harshman wrote:
> On 9/29/16 8:06 AM, Peter Nyikos wrote:
> > This is a reply to a post by George Kaplan on the rapidly expanding
> > thread [over 800 posts now],
> > Subject: Re: Why cannot we see evolution happening today?
> >
> > On https://groups.google.com/d/msg/talk.origins/mq1OKVRyRp4/Xu-PjO8qAAAJ on Wednesday, September 28, 2016 at 10:00:03 PM UTC-4,gkaplan wrote:
> >> On 9/28/2016 6:11 PM, Peter Nyikos wrote:
> >>> On Wednesday, September 28, 2016 at 4:00:02 PM UTC-4, gkaplan wrote:
> >>>> On 9/28/2016 11:26 AM, John Harshman wrote:
> >>>>> On 9/28/16 10:58 AM, gkaplan wrote:
> >>>>>> On 9/28/2016 10:24 AM, John Harshman wrote:
> >>>>>>> On 9/28/16 9:49 AM, gkaplan wrote:
> >>>>>>>
> >
> > [I don't know who wrote the following:]
> >>>>>>>>> Why don't you calculate for us the probability that exactly 55 base
> >>>>>>>>> pairs would be deleted from *exactly* the same spot in the genome
> >>>>>>>>> among all these various primates, without common ancestry being
> >>>>>>>>> involved?
> >
> > I'm not sure what this is supposed to refer to. My guess is that
> > "the same spot" refers to the same *locus* -- in this case, the
> > same pseudogene -- judging from what comes next.
>
> Maybe you should stop responding to comments that are 7 or 8 layers
> deep.


I have little choice on that furiously fast-paced thread; I simply
cannot keep up with all the posts there. I've tried to get George
to slow down, but I deleted that part from my OP reply on this thread
to keep the topic focused.

By the way, do YOU know who wrote that bit? The attribution line
for it was already missing from every post by George on that page
of New Google Groups.

> That might improve your understanding. It refers to a single
> deletion of a 55-base piece of the pseudogene.

As I had surmised below. Maybe you should stop replying to
things in a post without reading the whole post first.

> >>>>>>>> Have you considered the source could be genetic engineering and
> >>>>>>>> artificial selection?
> >>>>>>>
> >>>>>>> Why would anyone do genetic engineering and artificial selection in
> >>>>>>> order to fix a useless pseudogene mutation in a collection of unrelated
> >>>>>>> species?
> >>>>>>
> >>>>>> What is being fixed?
> >>>>>>

> >>>>> "Fixed", in biology, refers to a particular feature increasing in
> >>>>> frequency in the population until it's found in 100% of the population.
> >>>>> In this case, a deletion in a pseudogene is fixed in almost all primate
> >>>>> species.
> >
> > It seems you are talking here about a deletion of a whole fragment of a
> > [pseudo]gene, George.
>
> That wasn't George, that was me.

Sorry about the mixup.


> > If so, which primates still have that fragment? And is it part of a
> > gene that codes for a protein in them?
>
> It's part of a pseudogene. If I recall, this happened in the common
> ancestor of all primates in which it's a pseudogene. But I could be wrong.

Why did you say "almost all primate species"? [And here I thought George
had made that claim!]

> > In which sub-clade of the order *Primates* have those 55 base pairs
> > gone missing?
> >
> > If you don't know the answers, George, someone reading this might
> > know them.

Looks like you don't know them either. Perhaps the unknown
person who wrote that bit about "all these various primates"
will come to our rescue.

George had written:
> >>>> Somehow you come to the conclusion that because a pseudogene exists in
> >>>> an organism that if genetic engineering was the instrument, that the
> >>>> goal was to create a pseudogene? Are you attempting to be facetious?
> >
> > [here is where I came in:]
> >
> >>> Why should he be? Don't you know that some pseudogenes, or
> >>> "junk DNA" as it is popularly called, have been found to have useful
> >>> functions, just not the function of coding for polypeptides.
> >>> This has been called a triumph of ID theory by some, who make
> >>> the claim that ID theory had predicted that a lot of "junk DNA"
> >>> did fulfill some useful design plan.

<snip of things to which you had nothing to say>

> >> From what I have read, a deleterious mutation can
> >> also kill the organism or lower its fitness.
> >
> > Yes, by the mutant allele coding for the wrong kind of protein.
> >
> > But perhaps mutated pseudogenes can have the same deleterious effects.
>
> It's conceivable. Any sequence might have a mutation that would produce
> some deleterious effect, even a sequence that's been evolving neutrally
> for millions of years.

Do you think that particular pseudogene was not having any effects
at all for something like 50 million years? In that case, what's
the big deal about a deity putting it in?

One way to test this hypothesis is to see how many alleles that
one pseudogene has accumulated. At minimum, I would expect
every single primate species to have alleles that are not found in
any other primate species.

> Unlikely, though. I think what George is talking
> about is a mutation that would have been deleterious in a functional
> protein but that's neutral in a pseudogene. He may not realize this.

I hope he weighs in on this thread.

> >> The way I was was using the term was specific to my argument regarding a
> >> completely new function being created with multiple beneficial mutations
> >> to do anything significant on the same locus.
> >
> > In that case, why did you keep talking about 55 base pairs missing from
> > that pseudogene?
>
> I don't think he's been able to engage with the idea of a 55-base deletion.

I would have thought he'd read enough genetics to find out about
deletions like that.

> >> In that context it does
> >> not impact my argument if the gene is disabled or if it kills the
> >> organism. That trajectory is ended.
> >
> > But the question at the beginning evidently does not have anything
> > to do with this. A huge bunch of trajectories are very active
> > despite the presence of the pseudogene.
>
> Yes, he is exceeding confused, and yet he seems to think that everyone
> else is confused.

I haven't seen much evidence of either, but then, I haven't been
able to read more than a small fraction of posts to this thread.

And, aren't you in the same boat?

Peter Nyikos
Professor, Department of Math. -- standard disclaimer --
U. of South Carolina
http://www.math.sc.edu/~nyikos/

gkaplan

unread,
Sep 29, 2016, 2:05:01 PM9/29/16
to talk-o...@moderators.isc.org
Yes, my point was that the 55 bp deletion was an off-topic/new topic
discussion in the thread that has nothing to do with my previous,
present and future comments on whether gene duplication is a good source
of new function in putative human evolution.

John Harshman

unread,
Sep 29, 2016, 2:10:01 PM9/29/16
to talk-o...@moderators.isc.org
Perhaps you should just reconcile yourself to the fact that you will
lose many opportunities to respond, and not try to recover those
opportunities so long after the fact.

> By the way, do YOU know who wrote that bit? The attribution line
> for it was already missing from every post by George on that page
> of New Google Groups.

Vince.

>> That might improve your understanding. It refers to a single
>> deletion of a 55-base piece of the pseudogene.
>
> As I had surmised below. Maybe you should stop replying to
> things in a post without reading the whole post first.

Perhaps. But that seemed as convenient a place to tell you as anywhere.
Was your ego damaged by the implicit slighting of your cleverness?

>>>>>>>>>> Have you considered the source could be genetic engineering and
>>>>>>>>>> artificial selection?
>>>>>>>>>
>>>>>>>>> Why would anyone do genetic engineering and artificial selection in
>>>>>>>>> order to fix a useless pseudogene mutation in a collection of unrelated
>>>>>>>>> species?
>>>>>>>>
>>>>>>>> What is being fixed?
>>>>>>>>
>
>>>>>>> "Fixed", in biology, refers to a particular feature increasing in
>>>>>>> frequency in the population until it's found in 100% of the population.
>>>>>>> In this case, a deletion in a pseudogene is fixed in almost all primate
>>>>>>> species.
>>>
>>> It seems you are talking here about a deletion of a whole fragment of a
>>> [pseudo]gene, George.
>>
>> That wasn't George, that was me.
>
> Sorry about the mixup.

De nada. But if you stopped replying so many layers deep, you wouldn't
make such mistakes. Another benefit!

>>> If so, which primates still have that fragment? And is it part of a
>>> gene that codes for a protein in them?
>>
>> It's part of a pseudogene. If I recall, this happened in the common
>> ancestor of all primates in which it's a pseudogene. But I could be wrong.
>
> Why did you say "almost all primate species"? [And here I thought George
> had made that claim!]

Because I had misremembered. I've already posted a correction, but just
see this reference:

https://www.ncbi.nlm.nih.gov/pubmed/16102985

>>> In which sub-clade of the order *Primates* have those 55 base pairs
>>> gone missing?
>>>
>>> If you don't know the answers, George, someone reading this might
>>> know them.
>
> Looks like you don't know them either. Perhaps the unknown
> person who wrote that bit about "all these various primates"
> will come to our rescue.

Again, not an unknown person. Vince Maycock. Or you could try googling,
which is what I did. I've spoonfed the facts to you in another post in
this thread.

> George had written:
>>>>>> Somehow you come to the conclusion that because a pseudogene exists in
>>>>>> an organism that if genetic engineering was the instrument, that the
>>>>>> goal was to create a pseudogene? Are you attempting to be facetious?
>>>
>>> [here is where I came in:]
>>>
>>>>> Why should he be? Don't you know that some pseudogenes, or
>>>>> "junk DNA" as it is popularly called, have been found to have useful
>>>>> functions, just not the function of coding for polypeptides.
>>>>> This has been called a triumph of ID theory by some, who make
>>>>> the claim that ID theory had predicted that a lot of "junk DNA"
>>>>> did fulfill some useful design plan.
>
> <snip of things to which you had nothing to say>
>
>>>> From what I have read, a deleterious mutation can
>>>> also kill the organism or lower its fitness.
>>>
>>> Yes, by the mutant allele coding for the wrong kind of protein.
>>>
>>> But perhaps mutated pseudogenes can have the same deleterious effects.
>>
>> It's conceivable. Any sequence might have a mutation that would produce
>> some deleterious effect, even a sequence that's been evolving neutrally
>> for millions of years.
>
> Do you think that particular pseudogene was not having any effects
> at all for something like 50 million years? In that case, what's
> the big deal about a deity putting it in?

50 million years turns out to be too long, probably. In fact it isn't
clear, but the duplication could be as old as 30 million years or as
recent as 8. And the pseudogenization almost as old as 8 or as recent as
5.8. Now, I would presume that the duplicate has been evolving neutrally
ever since the duplication, since the original would compensate for any
deactivating mutation in the copy.

The pseudogene in humans is interesting in that recombination between
the pseudogene and the functional copy is known to result in genetic
disease in humans, possibly exacerbated by certain mutations in the
pseudogene. So there's an example of a deleterious mutation happening in
a functionless sequence. I would expect that there is now some selection
for deletion, i.e. if another big deletion happened in the pseudogene,
recombination would be less likely to occur.

Why would a deity put in a useless, or at best a pointless extra, copy
of a functional gene?

> One way to test this hypothesis is to see how many alleles that
> one pseudogene has accumulated. At minimum, I would expect
> every single primate species to have alleles that are not found in
> any other primate species.

What hypothesis? The one about neutral evolution for 50 million years or
the one about a deity? I don't think you're talking about allelic
variation here, really, but about divergence among species. You mean to
consider whether the pseudogene has accumulated changes at the rate
expected from neutral evolution.

Within the limited sample, this seems to be the case. Again, that sample
is human, chimp, gorilla, orangutan.

>> Unlikely, though. I think what George is talking
>> about is a mutation that would have been deleterious in a functional
>> protein but that's neutral in a pseudogene. He may not realize this.
>
> I hope he weighs in on this thread.
>
>>>> The way I was was using the term was specific to my argument regarding a
>>>> completely new function being created with multiple beneficial mutations
>>>> to do anything significant on the same locus.
>>>
>>> In that case, why did you keep talking about 55 base pairs missing from
>>> that pseudogene?
>>
>> I don't think he's been able to engage with the idea of a 55-base deletion.
>
> I would have thought he'd read enough genetics to find out about
> deletions like that.

He reads only to locate bits he can object to. This tends to impede
understanding of the main text.

>>>> In that context it does
>>>> not impact my argument if the gene is disabled or if it kills the
>>>> organism. That trajectory is ended.
>>>
>>> But the question at the beginning evidently does not have anything
>>> to do with this. A huge bunch of trajectories are very active
>>> despite the presence of the pseudogene.
>>
>> Yes, he is exceeding confused, and yet he seems to think that everyone
>> else is confused.
>
> I haven't seen much evidence of either, but then, I haven't been
> able to read more than a small fraction of posts to this thread.
>
> And, aren't you in the same boat?

Nope.

Peter Nyikos

unread,
Sep 29, 2016, 2:20:02 PM9/29/16
to talk-o...@moderators.isc.org
On Thursday, September 29, 2016 at 11:55:01 AM UTC-4, gkaplan wrote:
> On 9/29/2016 8:06 AM, Peter Nyikos wrote:
> > This is a reply to a post by George Kaplan on the rapidly expanding
> > thread [over 800 posts now],
> > Subject: Re: Why cannot we see evolution happening today?
> >
> > On https://groups.google.com/d/msg/talk.origins/mq1OKVRyRp4/Xu-PjO8qAAAJ on Wednesday, September 28, 2016 at 10:00:03 PM UTC-4,gkaplan wrote:
> >> On 9/28/2016 6:11 PM, Peter Nyikos wrote:
> >>> On Wednesday, September 28, 2016 at 4:00:02 PM UTC-4, gkaplan wrote:
> >>>> On 9/28/2016 11:26 AM, John Harshman wrote:
> >>>>> On 9/28/16 10:58 AM, gkaplan wrote:
> >>>>>> On 9/28/2016 10:24 AM, John Harshman wrote:
> >>>>>>> On 9/28/16 9:49 AM, gkaplan wrote:
> >>>>>>>
> >
> > [I don't know who wrote the following:]
> >>>>>>>>> Why don't you calculate for us the probability that exactly 55 base
> >>>>>>>>> pairs would be deleted from *exactly* the same spot in the genome
> >>>>>>>>> among all these various primates, without common ancestry being
> >>>>>>>>> involved?
> >
> > I'm not sure what this is supposed to refer to. My guess is that
> > "the same spot" refers to the same *locus* -- in this case, the
> > same pseudogene -- judging from what comes next.
>
> I believe this is John. The argument as I understand it, is that since
> Chimps and Gorillas and Humans show the same 55 bp deletion, that this
> proves common ancestry.

John seems to be saying that Vince Maycock wrote the above. The style
does look like Maycock's, for sure.

By the way, only some of the more ignorant participants on that
"parent thread" use the word "proves" in the way you suggest. Did
you encounter this argument somewhere outside of talk.origins?


> My response (in so many words) is that if a genetic engineer took the
> DNA from one organism and added a change, that what is seen would be
> consistent with this possibility. In other words, common ancestry is not
> the only possibility.

IMO, that would be a no-brainer if the pseudogene were restricted to
Hominidae, which now includes the orangutan. John seems to be stretching
his militant atheism to suggest otherwise, but I will address him on that
if you don't.


I had suggested one avenue to you below:

<snip for focus>

> >>>> Somehow you come to the conclusion that because a pseudogene exists in
> >>>> an organism that if genetic engineering was the instrument, that the
> >>>> goal was to create a pseudogene? Are you attempting to be facetious?
> >
> > [here is where I came in:]
> >
> >>> Why should he be? Don't you know that some pseudogenes, or
> >>> "junk DNA" as it is popularly called, have been found to have useful
> >>> functions, just not the function of coding for polypeptides.
> >>> This has been called a triumph of ID theory by some, who make
> >>> the claim that ID theory had predicted that a lot of "junk DNA"
> >>> did fulfill some useful design plan.

Perhaps you could take a look at the ID arguments for usefulness
of some pseudogenes. [The Discovery Institute would be a good place
to start.]

<snip for focus>

> >> In that context it does
> >> not impact my argument if the gene is disabled or if it kills the
> >> organism. That trajectory is ended.
> >
> > But the question at the beginning evidently does not have anything
> > to do with this. A huge bunch of trajectories are very active
> > despite the presence of the pseudogene.

No comment on this?

Anyway, I'm glad you found this thread so soon, George.

Alan Kleinman MD PhD

unread,
Sep 29, 2016, 3:35:01 PM9/29/16
to talk-o...@moderators.isc.org
Peter, explain to us why the evolution of a pseudogene would be any different than the evolution of any other gene. The mathematics that governs any evolutionary trajectory is simply a set of joint binomial probability problems. And the way that natural selection improves the probability for the next binomial probability problem in that set is by amplification of the particular variant. Only if a mutation occurs in that pseudogene that gives beneficial function improving that members ability to replicate will help that variant on its particular evolutionary trajectory.

John Harshman

unread,
Sep 29, 2016, 3:45:01 PM9/29/16
to talk-o...@moderators.isc.org
Pseudogenes (or at least the vast majority of them) don't have
"beneficial function", hence no selection. That's why they're
pseudogenes instead of genes. The only thing that would "help that
variant" is chance, i.e. drift. Would you agree that drift can happen
efficiently (as efficiently as it ever happens) in many loci at once?

Vincent Maycock

unread,
Sep 29, 2016, 4:15:01 PM9/29/16
to talk-o...@moderators.isc.org
On Thu, 29 Sep 2016 08:53:12 -0700, gkaplan <georg....@gmail.com>
wrote:

>On 9/29/2016 8:06 AM, Peter Nyikos wrote:
>> This is a reply to a post by George Kaplan on the rapidly expanding
>> thread [over 800 posts now],
>> Subject: Re: Why cannot we see evolution happening today?
>>
>> On https://groups.google.com/d/msg/talk.origins/mq1OKVRyRp4/Xu-PjO8qAAAJ on Wednesday, September 28, 2016 at 10:00:03 PM UTC-4,gkaplan wrote:
>>> On 9/28/2016 6:11 PM, Peter Nyikos wrote:
>>>> On Wednesday, September 28, 2016 at 4:00:02 PM UTC-4, gkaplan wrote:
>>>>> On 9/28/2016 11:26 AM, John Harshman wrote:
>>>>>> On 9/28/16 10:58 AM, gkaplan wrote:
>>>>>>> On 9/28/2016 10:24 AM, John Harshman wrote:
>>>>>>>> On 9/28/16 9:49 AM, gkaplan wrote:
>>>>>>>>
>>
>> [I don't know who wrote the following:]

That was me.

>>>>>>>>>> Why don't you calculate for us the probability that exactly 55 base
>>>>>>>>>> pairs would be deleted from *exactly* the same spot in the genome
>>>>>>>>>> among all these various primates, without common ancestry being
>>>>>>>>>> involved?
>>
>> I'm not sure what this is supposed to refer to. My guess is that
>> "the same spot" refers to the same *locus* -- in this case, the
>> same pseudogene -- judging from what comes next.
>
>I believe this is John. The argument as I understand it, is that since
>Chimps and Gorillas and Humans show the same 55 bp deletion, that this
>proves common ancestry.

I said at the same place (locus -- but beyond being in the same
pseudogene, it's at the*same spot* in the pseudogene).

>My response (in so many words) is that if a genetic engineer took the
>DNA from one organism and added a change, that what is seen would be
>consistent with this possibility. In other words, common ancestry is not
>the only possibility.

Why would your Genetic Engineer (capitalized to reflect who I think
you think this genetic engineer is) choose 55 base pairs to delete,
*to no effect*?

(Remember that pseudogenes have no function related to their sequence
-- John mentioned this as well, but you seem to have drifted off from
what he was telling you.)

What's the significance of 55, and why *that* spot in the genome?

Creationism (or Intelligent Design Creationism, whichever political
movement you identify with) can't answer these questions, while common
ancestry can.

gkaplan

unread,
Sep 29, 2016, 4:45:01 PM9/29/16
to talk-o...@moderators.isc.org
I never claimed the deletions were performed by a genetic engineer.
Don't you believe in mutations?

Alan Kleinman MD PhD

unread,
Sep 29, 2016, 4:55:00 PM9/29/16
to talk-o...@moderators.isc.org
Drift occurs when selection is random. And random selection can create population bottlenecks. But what is efficient in this case?

Burkhard

unread,
Sep 29, 2016, 5:40:00 PM9/29/16
to talk-o...@moderators.isc.org
Sure. We do reuse parts quite often between designs, if a) we are under
resource constraints (time, money etc) and/or lazy and b) want the two
things to do roughly the same thing anyway.

So independent evidence for an overworked, underfunded and slightly lazy
designer who had the same plan for humans and chimps would be evidence
that favours your hypothesis over the evolutionary alternative.

One could then explore if this new theory is productive, that is if it
leads us to new discoveries and observations.

John Harshman

unread,
Sep 29, 2016, 5:50:01 PM9/29/16
to talk-o...@moderators.isc.org
Your use of non-standard language makes it difficult to communicate with
you. Why do you use it? "Random selection" makes no sense as the words
are used in biology. Possibly we have the same problem with
"bottleneck". A bottleneck occurs if the population is reduced to a very
small number of individuals. Drift doesn't cause bottlenecks. Fixation
can happen without a bottleneck. Perhaps you are thinking of
coalescence? But that isn't a bottleneck either. Very confusing.

Would you agree that drift can happen in many loci at once? Would you
agree that the rate of fixation of a mutation at one locus does not
affect the rate of fixation of a mutation at another locus?

Alan Kleinman MD PhD

unread,
Sep 29, 2016, 7:15:00 PM9/29/16
to talk-o...@moderators.isc.org
You don't understand how drift works. Drift can create bottlenecks. A couple years ago Greg Guarino and several other debators showed using Thomas Schneider's EV computer simulation of rmns program how drift (random selection) creates bottlenecks. An easy way to understand how random selection creates a bottleneck, consider a deck of cards. Shuffle the deck and then randomly remove 26 cards. Of the remaining 26 cards, double each card. So if the Ace of hearts was removed, no more Ace of hearts in the deck. If the Ace of spades is still in the deck, you now have two aces of spades and so on. Shuffle the deck again and then randomly remove 26 cards and double the remaining cards. So if both Ace spades were still in the deck, you now have 4 aces of spades. If 1 of 2 kings of clubs were removed, you still have 2 kings of clubs after the doubling. If you continue this process over and over, you will end up with a single card repeated 52 times. If that card is the Ace of Spades, it was the bottleneck in the population.

Peter Nyikos

unread,
Sep 29, 2016, 8:30:00 PM9/29/16
to talk-o...@moderators.isc.org
Yes, that became clear from what Harshman wrote later on.

> >My response (in so many words) is that if a genetic engineer took the
> >DNA from one organism and added a change, that what is seen would be
> >consistent with this possibility. In other words, common ancestry is not
> >the only possibility.
>
> Why would your Genetic Engineer (capitalized to reflect who I think
> you think this genetic engineer is) choose 55 base pairs to delete,
> *to no effect*?
>
> (Remember that pseudogenes have no function related to their sequence
> -- John mentioned this as well, but you seem to have drifted off from
> what he was telling you.)

You are your usual sophomoric self here, Vince. If you hadn't triumphantly
exited before reading that some pseudogenes HAVE been shown to have
uses, just not the use of coding for proteins, you MIGHT not have
made this stupid comment.

> What's the significance of 55, and why *that* spot in the genome?
>
> Creationism (or Intelligent Design Creationism, whichever political
> movement you identify with)

Creationism isn't exclusively political, and ID theorists are no more
political than evolutionary theorists and paleontologists. [There are some real political animals among the latter, including Donald Prothero and Christine Janis. And so I allow for some political animals among ID
theorists.]

And the ID movement is NOT a branch of creationism, and no amount of
your parroting of propaganda by the likes of Laurence A. Moran is going to
change that.

> can't answer these questions, while common
> ancestry can.

Your failure to read to the end of the post to which you are replying is
really showing here.

Or did you just decide to indulge in a bunch of trolling in this post?

<snip to get to the relevant part>

> >>>>> Somehow you come to the conclusion that because a pseudogene exists in
> >>>>> an organism that if genetic engineering was the instrument, that the
> >>>>> goal was to create a pseudogene? Are you attempting to be facetious?
> >>
> >> [here is where I came in:]
> >>
> >>>> Why should he be? Don't you know that some pseudogenes, or
> >>>> "junk DNA" as it is popularly called, have been found to have useful
> >>>> functions, just not the function of coding for polypeptides.
> >>>> This has been called a triumph of ID theory by some, who make
> >>>> the claim that ID theory had predicted that a lot of "junk DNA"
> >>>> did fulfill some useful design plan.

<snip to get to some related information>

> >>> From what I have read, a deleterious mutation can
> >>> also kill the organism or lower its fitness.
> >>
> >> Yes, by the mutant allele coding for the wrong kind of protein.
> >>
> >> But perhaps mutated pseudogenes can have the same deleterious effects.

You are too ignorant about pseudogenes to comment on this,
aren't you, Vince?

Peter Nyikos
Professor, Dept. of Mathematics -- standard disclaimer--
University of South Carolina
http://people.math.sc.edu/nyikos/

PS Did you ever see the following reply by me to you?

https://groups.google.com/d/msg/talk.origins/mq1OKVRyRp4/sKV3vvrHBgAJ
Subject: Re: Why cannot we see evolution happening today?
Date: Mon, 26 Sep 2016 17:53:26 -0700 (PDT)
Message-ID: <25c9efa5-32fe-447f...@googlegroups.com>

John Harshman

unread,
Sep 29, 2016, 8:40:00 PM9/29/16
to talk-o...@moderators.isc.org
Yes, I understand how drift works. What you have described is
coalescence, not a bottleneck. Note that the population remains
constant. Now imagine that a genome consists of a bunch of decks, all
undergoing that same process. Each deck will eventually end up with a
single card replicated 52 times, but each deck will likely have a
different card, tracing back to different decks originating at different
times. Thus, in a species in which different loci can be inherited by
different pathways (i.e. sex, recombination), lots of different alleles
at different loci can drift to fixation simultaneously.

John Harshman

unread,
Sep 29, 2016, 9:25:00 PM9/29/16
to talk-o...@moderators.isc.org
To be fair to Vince, it's a negligible number and does not include the
one we're talking about right now. To a first approximation, pseudogenes
have no function.

>> What's the significance of 55, and why *that* spot in the genome?
>>
>> Creationism (or Intelligent Design Creationism, whichever political
>> movement you identify with)
>
> Creationism isn't exclusively political, and ID theorists are no more
> political than evolutionary theorists and paleontologists. [There are some real political animals among the latter, including Donald Prothero and Christine Janis. And so I allow for some political animals among ID
> theorists.]
>
> And the ID movement is NOT a branch of creationism, and no amount of
> your parroting of propaganda by the likes of Laurence A. Moran is going to
> change that.

You live in your own little world and seem not to be familiar with the
actual ID movement. "Cdesign proponentsists" says it all, really.


Alan Kleinman MD PhD

unread,
Sep 29, 2016, 9:25:00 PM9/29/16
to talk-o...@moderators.isc.org
I think of coalescence in different terms than you. I see coalescence as two (or more) things joining together. The card deck analogy is demonstrating that all duplications ultimately came from a single card where that original card represents the "bottleneck". It is true that you don't know in advance which card will be the progenitor of all the cards in the deck but that is random selection.

Peter Nyikos

unread,
Sep 29, 2016, 9:55:00 PM9/29/16
to talk-o...@moderators.isc.org
On Thursday, September 29, 2016 at 2:10:01 PM UTC-4, John Harshman wrote:
> On 9/29/16 10:44 AM, Peter Nyikos wrote:
> > On Thursday, September 29, 2016 at 11:50:02 AM UTC-4, John Harshman wrote:
> >> On 9/29/16 8:06 AM, Peter Nyikos wrote:
> >>> This is a reply to a post by George Kaplan on the rapidly expanding
> >>> thread [over 800 posts now],

In one day, it has gone from 800 to 860, and the evening is yet young.
You keep trying to stir up a tempest in a teapot based on my
not knowing who wrote the first thing I quoted in my OP. Stop it.

> >>>>>>>>>> Have you considered the source could be genetic engineering and
> >>>>>>>>>> artificial selection?
> >>>>>>>>>
> >>>>>>>>> Why would anyone do genetic engineering and artificial selection in
> >>>>>>>>> order to fix a useless pseudogene mutation in a collection of unrelated
> >>>>>>>>> species?
> >>>>>>>>
> >>>>>>>> What is being fixed?
> >>>>>>>>
> >
> >>>>>>> "Fixed", in biology, refers to a particular feature increasing in
> >>>>>>> frequency in the population until it's found in 100% of the population.
> >>>>>>> In this case, a deletion in a pseudogene is fixed in almost all primate
> >>>>>>> species.
> >>>
> >>> It seems you are talking here about a deletion of a whole fragment of a
> >>> [pseudo]gene, George.
> >>
> >> That wasn't George, that was me.
> >
> > Sorry about the mixup.
>
> De nada. But if you stopped replying so many layers deep, you wouldn't
> make such mistakes. Another benefit!

I'm willing to risk such trivial mistakes. Would that all our mistakes
were this trivial!

> >>> If so, which primates still have that fragment? And is it part of a
> >>> gene that codes for a protein in them?
> >>
> >> It's part of a pseudogene. If I recall, this happened in the common
> >> ancestor of all primates in which it's a pseudogene. But I could be wrong.
> >
> > Why did you say "almost all primate species"? [And here I thought George
> > had made that claim!]
>
> Because I had misremembered. I've already posted a correction, but just
> see this reference:
>
> https://www.ncbi.nlm.nih.gov/pubmed/16102985

Very interesting. Thanks.

> >>> In which sub-clade of the order *Primates* have those 55 base pairs
> >>> gone missing?
> >>>
> >>> If you don't know the answers, George, someone reading this might
> >>> know them.
> >
> > Looks like you don't know them either. Perhaps the unknown
> > person who wrote that bit about "all these various primates"
> > will come to our rescue.
>
> Again, not an unknown person. Vince Maycock. Or you could try googling,
> which is what I did. I've spoonfed the facts to you in another post in
> this thread.

This use of "spoonfed" is uncalled for. The link you gave does the
job better than your "spoonfeeding".

<snip to finally get to something constructive by you>

> > Do you think that particular pseudogene was not having any effects
> > at all for something like 50 million years? In that case, what's
> > the big deal about a deity putting it in?
>
> 50 million years turns out to be too long, probably. In fact it isn't
> clear, but the duplication could be as old as 30 million years or as
> recent as 8. And the pseudogenization almost as old as 8 or as recent as
> 5.8. Now, I would presume that the duplicate has been evolving neutrally
> ever since the duplication, since the original would compensate for any
> deactivating mutation in the copy.

That last sentence is sheer speculation, unless there is something in
the linked article to support it. [I've only read the abstract, the
rest being paywalled; but I can get at it at my office tomorrow.]

> The pseudogene in humans is interesting in that recombination between
> the pseudogene and the functional copy is known to result in genetic
> disease in humans, possibly exacerbated by certain mutations in the
> pseudogene. So there's an example of a deleterious mutation happening in
> a functionless sequence. I would expect that there is now some selection
> for deletion, i.e. if another big deletion happened in the pseudogene,
> recombination would be less likely to occur.
>
> Why would a deity put in a useless, or at best a pointless extra, copy
> of a functional gene?

Unfortunately, George shows no inclination to look for
information about the usefulness of some pseudogenes, and I don't have
the time for it, for another month at least. So we'll have to be
stalemated about the alleged uselessness of this particular pseudogene.

> > One way to test this hypothesis is to see how many alleles that
> > one pseudogene has accumulated. At minimum, I would expect
> > every single primate species to have alleles that are not found in
> > any other primate species.
>
> What hypothesis? The one about neutral evolution for 50 million years or
> the one about a deity?

Don't be silly: of course I meant the former, except that now we see
that it is a much shorter time, and we only have four or five species
to deal with.

Wait, make that six. By population genetic standards, the mountain
gorilla is a separate species from the lowland gorilla.

[For that matter, those standards probably give us at least a dozen
different species [= reproductively isolated tribes] of human beings on
various hilltops in New Guinea, but that issue would call for a separate
thread.]

> I don't think you're talking about allelic
> variation here, really, but about divergence among species.

The two go hand in hand.

> You mean to
> consider whether the pseudogene has accumulated changes at the rate
> expected from neutral evolution.

Totally neutral, as befits a totally non-functioning pseudogene.
[And I don't mean just non-protein-coding, which is what all too many
talk.origins participants think "non-functional" means.]

> Within the limited sample, this seems to be the case. Again, that sample
> is human, chimp, gorilla, orangutan.

Could you tell from the complete article whether the bonobo was
also sampled?

<snip unsupported comments about George, with the last one left in for context:>

> >> Yes, he is exceeding confused, and yet he seems to think that everyone
> >> else is confused.
> >
> > I haven't seen much evidence of either, but then, I haven't been
> > able to read more than a small fraction of posts to this thread.
> >
> > And, aren't you in the same boat?
>
> Nope.

Are you suggesting that you read everything on that "parent" thread?

If so, I wonder whether you'll deign to address any of what I've
written to Kleinman these last two days about Sphenosuchia
and other evolutionary matters.

Peter Nyikos
Professor, Dept. of Mathematics -- standard disclaimer --
U. of S. Carolina
http://people.math.sc.edu/nyikos

Peter Nyikos

unread,
Sep 29, 2016, 10:20:00 PM9/29/16
to talk-o...@moderators.isc.org
Your biased idea of "fair" is a knife that only cuts one way.

> it's a negligible number and does not include the
> one we're talking about right now.

How do you know that? Did you read the actual, paywalled article?

And what percentage is "negligible"? 10%? 1%?

> To a first approximation, pseudogenes
> have no function.

Spoken like a true propagandist, spin-doctoring reality to make
the gratuitous-insult-delivering Maycock look reasonable.

> >> What's the significance of 55, and why *that* spot in the genome?
> >>
> >> Creationism (or Intelligent Design Creationism, whichever political
> >> movement you identify with)
> >
> > Creationism isn't exclusively political, and ID theorists are no more
> > political than evolutionary theorists and paleontologists. [There are some
> > real political animals among the latter, including Donald Prothero and Christine Janis.
> > And so I allow for some political animals among ID theorists.]

> > And the ID movement is NOT a branch of creationism, and no amount of
> > your parroting of propaganda by the likes of Laurence A. Moran is going to
> > change that.
>
> You live in your own little world and seem not to be familiar with the
> actual ID movement. "Cdesign proponentsists" says it all, really.

You need to get over your decade+ long love affair with that word, and
read what ID theorists actually write, especially Behe.

That's a tall order for a polemicist like you, so I'll leave you with
this middle-school-level approximation to what I mean:

Creationism : "proof" :: ID theory : evidence

Even that is a tall order, because it's hard for you to resist the urge
to pander to people who use "no evidence" to mean "no evidence sufficient
to leave me at a loss for an answer."

I do acknowledge that their evidence is meager except in the eyes of
most believing Christians, just as your evidence for God's
nonexistence is meager except in the eyes of your fellow militant atheists.

Peter Nyikos
Professor of Mathematics -- standard disclaimer
University of South Carolina

Peter Nyikos

unread,
Sep 29, 2016, 10:35:01 PM9/29/16
to talk-o...@moderators.isc.org
This is a reply to George Kaplan on the "parent" thread.

On Thursday, September 29, 2016 at 7:00:01 PM UTC-4, gkaplan wrote:
> On 9/29/2016 2:51 PM, John Harshman wrote:
> > On 9/29/16 1:35 PM, gkaplan wrote:
> >> On 9/29/2016 1:21 PM, John Harshman wrote:
> >>> On 9/29/16 1:14 PM, gkaplan wrote:
> >>>> On 9/29/2016 1:03 PM, John Harshman wrote:
> >>>>> On 9/29/16 12:49 PM, gkaplan wrote:
> >>>>>> On 9/29/2016 12:38 PM, John Harshman wrote:
> >>>>>>> On 9/29/16 12:10 PM, gkaplan wrote:
> >>>>>>>> On 9/29/2016 11:50 AM, Mark Isaak wrote:
> >>>>>>>>> On 9/28/16 5:56 PM, gkaplan wrote:
> >>>>>>>>>> On 9/28/2016 5:26 PM, Mark Isaak wrote:
> >>>>>>>>>>> On 9/27/16 1:18 PM, gkaplan wrote:
> >>>>>>>>>>>> On 9/27/2016 12:59 PM, Greg Guarino wrote:
> >>>>>>>>>>>>> On 9/27/2016 1:47 PM, gkaplan wrote:

[I believe it was Norman who wrote:]
> >>>>>>>>>>>>>>> You say it lowers the probability of this actually
> >>>>>>>>>>>>>>> happening.
> >>>>>>>>>>>>>>> The
> >>>>>>>>>>>>>>> fact is that it does happen. And it happens a lot
> >>>>>>>>>>>>>>> because we
> >>>>>>>>>>>>>>> see an
> >>>>>>>>>>>>>>> awful lot of gene families that result.
> >>>>>>>>>>>>>>
> >>>>>>>>>>>>>> Your logic is circular. Basically it is that "it happens a
> >>>>>>>>>>>>>> lot
> >>>>>>>>>>>>>> because
> >>>>>>>>>>>>>> we see it a lot". That assumes your explanation is correct.

Here is your accusation of circularity I've been talking about, George.

We see a lot of what looks like gene duplication and subsequent
divergence, and so "It happens a lot." This "and so" is based on
cladistic analyses of various genes which makes them look like they
go back to various common ancestors. But the cladistic analysis only
shows that "It happens a lot" because of an *a priori* assumption
that the branching pattern is recapturing actual evolution.

This is what I was referring to when I wrote:


But people like Harshman and Norman will indulge in the circular
argument that SINCE WE KNOW common descent took place, the trees DO
illustrate common descent. The nodes (forks) in the tree are KNOWN to
represent hypothetical (yes, they use that word!) Last Common Ancestors.

The above applies equally well to animals and to genes.

<snip to get to a comment by you two posts later>

> >>>>>>>>>>>> My textbook defines gene family as: An ensemble of similar
> >>>>>>>>>>>> genes
> >>>>>>>>>>>> formed
> >>>>>>>>>>>> when a precursor gene undergoes duplication, and then the
> >>>>>>>>>>>> precursor or
> >>>>>>>>>>>> derivative genes undergo further duplications to produce
> >>>>>>>>>>>> multiple
> >>>>>>>>>>>> copies
> >>>>>>>>>>>> within the genome. (Chapter 13) Bergstrom, Carl T.; Dugatkin,
> >>>>>>>>>>>> Lee
> >>>>>>>>>>>> Alan.
> >>>>>>>>>>>> Evolution (Second Edition) (Page A-14). W. W. Norton & Company.

As you can see, that underlying assumption is at work. As it should
be, IMO. One should not invoke divine intervention haphazardly,
even if one believes in guided evolution. [I don't but I would assign a
probability at least 10% to there being SOME divinely guided evolution
if I was convinced of the existence of God. But while I would like it
very much if that were true, I only assign a probability of at most 10%
for that.]

> >>>>>>>>>> If someone took the DNA of a Chimpanzee and modified it to give
> >>>>>>>>>> it a
> >>>>>>>>>> completely new capability, how would that look different than
> >>>>>>>>>> common
> >>>>>>>>>> descent?

Now Mark Isaak sets an exceptionally high bar -- or so he probably
thinks:

> >>>>>>>>> Depends on the modification. If they took genes for cobra
> >>>>>>>>> venom or
> >>>>>>>>> firefly luciferase, then we would see a pattern inconsistent with
> >>>>>>>>> common
> >>>>>>>>> descent,

...unless one suspects lateral transfer of genes by retroviruses
transmitted by insect vectors. I'm sure Mark Isaak would latch on to
such a scenario for dear life if his setting-the-bar-high
ploy were to fail. And he'd justify it with something along the
lines of "Extraordinary claims require extraordinary evidence."

The majority of people here are that set against intelligent design,
even by natural means such as extraterrestrial species.

<snip to get to some words by Harshman>

> >>>>> If new features come about by angelic intervention in a lineage that
> >>>>> otherwise accumulates random deletions and such, that's theistic
> >>>>> evolution.
> >>>>
> >>>> So, if on the sixth creative day (not a 24 hour day) the Father told
> >>>> His Son to have the angel in charge of mammal development to take the
> >>>> DNA of an existing primate and make modifications so that the new
> >>>> organism had new features (we call this a chimpanzee today) that is
> >>>> theistic evolution?
> >>>
> >>> Sure. All you're doing is making angels the source of some mutations,
> >>> right? Of course this "existing primate", which is the source (common
> >>> ancestor) of both humans and chimps, is now extinct, and had experienced
> >>> quite a long time of evolution since it was engineered from the source
> >>> (again, common ancestor) of itself and gorillas. And so on.
> >>
> >> So, if the only mutations that occured from the time A was engineered
> >> till B was modified from it were neutral, and had no effect on
> >> phenotype, that would still be considered theistic evolution?
> >
> > Yes it would. Neutral evolution is still evolution. But you are
> > postulating mutations that affect phenotype; you're just claiming they
> > were put there by angels rather than natural processes. That's theistic
> > evolution.
>
> So where do you draw the line? If angels instead fabricated DNA without
> any [neutral] random mutations to make a chimpanzee, would that also be
> theistic evolution?

ANY evolution that involves the input of angels created and guided by God
counts as theistic evolution.

Why didn't you see this right away? Don't let Harshman's shenanigans
elsewhere fool you. He's quite sober and on target here.

> >>> (Love your hierarchical organization of heaven, in which creation has to
> >>> go through at least two layers of bureaucracy, by the way.)

Here, however, Harshman *seems* very naive about Christianity of
all kinds. Whether he is putting it on, and merely being disingenuous,
I cannot say for sure, but my suspicion is that he is.

Peter Nyikos
Professor, Dept. of Mathematics -- standard disclaimer--
U. of South Carolina
http://people.math.sc.edu/nyikos/

John Harshman

unread,
Sep 30, 2016, 1:05:00 AM9/30/16
to talk-o...@moderators.isc.org
On 9/29/16 7:32 PM, Peter Nyikos wrote:
> This is a reply to George Kaplan on the "parent" thread.
>
> On Thursday, September 29, 2016 at 7:00:01 PM UTC-4, gkaplan wrote:

>>>>>> So, if on the sixth creative day (not a 24 hour day) the Father told
>>>>>> His Son to have the angel in charge of mammal development to take the
>>>>>> DNA of an existing primate and make modifications so that the new
>>>>>> organism had new features (we call this a chimpanzee today) that is
>>>>>> theistic evolution?

>>>>> (Love your hierarchical organization of heaven, in which creation has to
>>>>> go through at least two layers of bureaucracy, by the way.)
>
> Here, however, Harshman *seems* very naive about Christianity of
> all kinds. Whether he is putting it on, and merely being disingenuous,
> I cannot say for sure, but my suspicion is that he is.

Please explain.

John Harshman

unread,
Sep 30, 2016, 1:14:59 AM9/30/16
to talk-o...@moderators.isc.org
On 9/29/16 6:52 PM, Peter Nyikos wrote:
> On Thursday, September 29, 2016 at 2:10:01 PM UTC-4, John Harshman wrote:
>> On 9/29/16 10:44 AM, Peter Nyikos wrote:

>>> Do you think that particular pseudogene was not having any effects
>>> at all for something like 50 million years? In that case, what's
>>> the big deal about a deity putting it in?
>>
>> 50 million years turns out to be too long, probably. In fact it isn't
>> clear, but the duplication could be as old as 30 million years or as
>> recent as 8. And the pseudogenization almost as old as 8 or as recent as
>> 5.8. Now, I would presume that the duplicate has been evolving neutrally
>> ever since the duplication, since the original would compensate for any
>> deactivating mutation in the copy.
>
> That last sentence is sheer speculation, unless there is something in
> the linked article to support it. [I've only read the abstract, the
> rest being paywalled; but I can get at it at my office tomorrow.]

No, it's just how duplicate genes work. Since there is the original copy
still serving the original function, the new copy is a spare, and
deactivation can't be deleterious.

>> The pseudogene in humans is interesting in that recombination between
>> the pseudogene and the functional copy is known to result in genetic
>> disease in humans, possibly exacerbated by certain mutations in the
>> pseudogene. So there's an example of a deleterious mutation happening in
>> a functionless sequence. I would expect that there is now some selection
>> for deletion, i.e. if another big deletion happened in the pseudogene,
>> recombination would be less likely to occur.
>>
>> Why would a deity put in a useless, or at best a pointless extra, copy
>> of a functional gene?
>
> Unfortunately, George shows no inclination to look for
> information about the usefulness of some pseudogenes, and I don't have
> the time for it, for another month at least. So we'll have to be
> stalemated about the alleged uselessness of this particular pseudogene.

No we won't. This particular pseudogene is useless. At the very least it
should be assumed useless until there's good evidence otherwise, as most
pseudogenes are useless. And in fact your test has more or less been
performed. It's evolving much faster than conserved sequences.

>>> One way to test this hypothesis is to see how many alleles that
>>> one pseudogene has accumulated. At minimum, I would expect
>>> every single primate species to have alleles that are not found in
>>> any other primate species.
>>
>> What hypothesis? The one about neutral evolution for 50 million years or
>> the one about a deity?
>
> Don't be silly: of course I meant the former, except that now we see
> that it is a much shorter time, and we only have four or five species
> to deal with.
>
> Wait, make that six. By population genetic standards, the mountain
> gorilla is a separate species from the lowland gorilla.

No, still 5. There is sequence only for one of those, and in fact one
individual.

> [For that matter, those standards probably give us at least a dozen
> different species [= reproductively isolated tribes] of human beings on
> various hilltops in New Guinea, but that issue would call for a separate
> thread.]

It's a silly issue. There are no isolated tribes.

>> I don't think you're talking about allelic
>> variation here, really, but about divergence among species.
>
> The two go hand in hand.

They do only if certain assumptions about population size stability are
met, which they are not. Humans, you have been told, have less allelic
diversity than most species, including chimps.

>> You mean to
>> consider whether the pseudogene has accumulated changes at the rate
>> expected from neutral evolution.
>
> Totally neutral, as befits a totally non-functioning pseudogene.
> [And I don't mean just non-protein-coding, which is what all too many
> talk.origins participants think "non-functional" means.]

Do you know of anyone who thinks that?

>> Within the limited sample, this seems to be the case. Again, that sample
>> is human, chimp, gorilla, orangutan.
>
> Could you tell from the complete article whether the bonobo was
> also sampled?

It was not. Five samples only.

> <snip unsupported comments about George, with the last one left in for context:>
>
>>>> Yes, he is exceeding confused, and yet he seems to think that everyone
>>>> else is confused.
>>>
>>> I haven't seen much evidence of either, but then, I haven't been
>>> able to read more than a small fraction of posts to this thread.
>>>
>>> And, aren't you in the same boat?
>>
>> Nope.
>
> Are you suggesting that you read everything on that "parent" thread?

I read enough to know what's being talked about.

> If so, I wonder whether you'll deign to address any of what I've
> written to Kleinman these last two days about Sphenosuchia
> and other evolutionary matters.

No, I'm just going to watch with amusement for a while while you fight
it out. Remember, Kleinman denies that most evolution can actually
happen. He doesn't care whether birds are not dinosaurs or not
thecodonts. They're not descended from anything but were created as birds.

John Harshman

unread,
Sep 30, 2016, 1:20:00 AM9/30/16
to talk-o...@moderators.isc.org
You still have no clue. Let me try again. In a clonal species there is a
single coalescent for the entire genome. In a recombining species there
can be a different coalescent for every base. Do you understand that?

Bill Rogers

unread,
Sep 30, 2016, 7:44:59 AM9/30/16
to talk-o...@moderators.isc.org

Vincent Maycock

unread,
Sep 30, 2016, 10:04:59 AM9/30/16
to talk-o...@moderators.isc.org
On Thu, 29 Sep 2016 17:24:51 -0700 (PDT), Peter Nyikos
<nyi...@bellsouth.net> wrote:


snip

>> (Remember that pseudogenes have no function related to their sequence
>> -- John mentioned this as well, but you seem to have drifted off from
>> what he was telling you.)
>
>You are your usual sophomoric self here, Vince. If you hadn't triumphantly
>exited before reading that some pseudogenes HAVE been shown to have
>uses, just not the use of coding for proteins, you MIGHT not have
>made this stupid comment.

I said "no function related to their sequence."

gkaplan

unread,
Sep 30, 2016, 11:09:59 AM9/30/16
to talk-o...@moderators.isc.org
> Peter Nyikos
> Professor, Dept. of Mathematics -- standard disclaimer--
> University of South Carolina
> http://people.math.sc.edu/nyikos/
>


Evolutionary direction of processed pseudogenes
By: Liu, Guoqing; Cui, Xiangjun; Li, Hong; et al.
SCIENCE CHINA-LIFE SCIENCES Volume: 59 Issue: 8 Pages: 839-849
Published: AUG 2016

While some pseudogenes have been reported to play important roles in
gene regulation, little is known about the possible relationship between
pseudogene functions and evolutionary process of pseudogenes, or about
the forces responsible for the pseudogene evolution. In this study, we
characterized human processed pseudogenes in terms of evolutionary
dynamics. Our results show that pseudogenes tend to evolve toward: lower
GC content, strong dinucleotide bias, reduced abundance of transcription
factor binding motifs and short palindromes, and decreased ability to
form nucleosomes. We explored possible evolutionary forces that shaped
the evolution pattern of pseudogenes, and concluded that mutations in
pseudogenes are likely determined, at least partially, by
neighbor-dependent mutational bias and recombination-associated selection.

---


Noise-induced multistability in the regulation of cancer by genes and
pseudogenes
By: Petrosyan, K. G.; Hu, Chin-Kun
JOURNAL OF CHEMICAL PHYSICS Volume: 145 Issue: 4 Article Number:
045102 Published: JUL 28 2016

By: Petrosyan, K. G.; Hu, Chin-Kun
JOURNAL OF CHEMICAL PHYSICS Volume: 145 Issue: 4 Article Number:
045102 Published: JUL 28 2016

We extend a previously introduced model of stochastic gene regulation of
cancer to a nonlinear case having both gene and pseudogene messenger
RNAs (mRNAs) self-regulated. The model consists of stochastic Boolean
genetic elements and possesses noise-induced multistability
(multimodality). We obtain analytical expressions for probabilities for
the case of constant but finite number of microRNA molecules which act
as a noise source for the competing gene and pseudogene mRNAs. The
probability distribution functions display both the global bistability
regime as well as even-odd number oscillations for a certain range of
model parameters. Statistical characteristics of the mRNA's level
fluctuations are evaluated. The obtained results of the extended model
advance our understanding of the process of stochastic gene and
pseudogene expressions that is crucial in regulation of cancer.
Published by AIP Publishing.

---


HMGA1-pseudogenes and cancer
By: De Martino, Marco; Forzati, Floriana; Arra, Claudio; et al.
ONCOTARGET Volume: 7 Issue: 19 Pages: 28724-28735 Published: MAY
10 2016

Abstract
Pseudogenes are DNA sequences with high homology to the corresponding
functional gene, but, because of the accumulation of various mutations,
they have lost their initial functions to code for proteins.
Consequently, pseudogenes have been considered until few years ago
dysfunctional relatives of the corresponding ancestral genes, and then
useless in the course of genome evolution. However, several studies have
recently established that pseudogenes are owners of key biological
functions. Indeed, some pseudogenes control the expression of functional
genes by competitively binding to the miRNAs, some of them generate
small interference RNAs to negatively modulate the expression of
functional genes, and some of them even encode functional mutated
proteins. Here, we concentrate our attention on the pseudogenes of the
HMGA1 gene, that codes for the HMGA1a and HMGA1b proteins having a
critical role in development and cancer progression. In this review, we
analyze the family of HMGA1 pseudogenes through three aspects:
classification, characterization, and their possible function and
involvement in cancer.

---

Pseudogenes: a novel source of trans-acting antisense RNAs.
By:Johnsson, Per; Morris, Kevin V; Grander, Dan
Methods in molecular biology (Clifton, N.J.)
Volume:1167 Pages:213-26
DOI:10.1007/978-1-4939-0835-6_14
Published:2014

While long thought to represent only "junk" DNA, several recent studies
support a functional role for pseudogenes. Several hundreds of
pseudogenes have been reported as transcribed into RNA in a large
variety of tissues and tumors. Most studies have focused on pseudogenes
expressed in the sense direction, but some reports suggest that
pseudogenes can be also transcribed as antisense RNAs (asRNAs). A few
examples of key regulatory genes, such as PTEN and OCT4, have in fact
been reported to be under the regulation of pseudogene-expressed asRNAs.
Here, we review what is known about pseudogene-expressed asRNAs and we
discuss the functional role that these transcripts may have in gene
regulation. Finally, we discuss some technical challenges when
characterising the function of pseudogene asRNAs.

---
Four products from Escherichia coli pseudogenes increase hydrogen production
By:Yusoff, MZM (Yusoff, Mohd Zulkhairi Mohd)[ 1,2,3,4 ] ; Hashiguchi, Y
(Hashiguchi, Yuya)[ 1 ] ; Maeda, T (Maeda, Toshinari)[ 1 ] ; Wood, TK
(Wood, Thomas K.)[ 2,3 ]
View ResearcherID and ORCID
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
Volume: 439 Issue: 4 Pages: 576-579
DOI: 10.1016/j.bbrc.2013.09.016
Published: OCT 4 2013



Pseudogenes are considered to be nonfunctional genes that lack a
physiological role. By screening 3985 Escherichia coli mutants using
chemochromic membranes, we found four pseudogenes involved in hydrogen
metabolism. Knockouts of pseudogenes ydfW and ypdJ had a defective
hydrogen phenotype on glucose and formate, respectively. Also, the
knockout of pseudogene yqiG formed hydrogen from formate but not from
glucose. For the yqiG mutant, 100% hydrogen recovery was obtained by the
complementation of YqiG via a plasmid. The knockout of pseudogene ylcE
showed hydrogen deficiency in minimal media which suggested that the
role of YlcE is associated with cell growth. Hence, the products of
these four pseudogenes play an important physiological role in hydrogen
production in E. coli. (C) 2013 The Authors. Published by Elsevier Inc.
All rights reserved.






John Harshman

unread,
Sep 30, 2016, 12:29:58 PM9/30/16
to talk-o...@moderators.isc.org
You really should listen to yourself some time.

>> it's a negligible number and does not include the
>> one we're talking about right now.
>
> How do you know that? Did you read the actual, paywalled article?
>
> And what percentage is "negligible"? 10%? 1%?

The common estimate for the number of human pseudogenes is around
20,000. A quick literature search finds various assays for functionality
(most commonly sequence conservation) have turned up only a "handful"
(quote from one of the surveys) of candidates, certainly less than 10.
So it may be on the order of 0.1% to 0.01%.

>> To a first approximation, pseudogenes
>> have no function.
>
> Spoken like a true propagandist, spin-doctoring reality to make
> the gratuitous-insult-delivering Maycock look reasonable.

You should probably try harder to separate his gratuitous insults from
his substantive claims. I know I try to do that with you.

>>>> What's the significance of 55, and why *that* spot in the genome?
>>>>
>>>> Creationism (or Intelligent Design Creationism, whichever political
>>>> movement you identify with)
>>>
>>> Creationism isn't exclusively political, and ID theorists are no more
>>> political than evolutionary theorists and paleontologists. [There are some
>>> real political animals among the latter, including Donald Prothero and Christine Janis.
>>> And so I allow for some political animals among ID theorists.]
>
>>> And the ID movement is NOT a branch of creationism, and no amount of
>>> your parroting of propaganda by the likes of Laurence A. Moran is going to
>>> change that.
>>
>> You live in your own little world and seem not to be familiar with the
>> actual ID movement. "Cdesign proponentsists" says it all, really.
>
> You need to get over your decade+ long love affair with that word, and
> read what ID theorists actually write, especially Behe.

Especially Behe or only Behe? That word is a stand-in for a whole lot of
explicit connections to creationism and religion. We've gone over some
of those before. Behe is the most circumspect of them all, but prominent
IDers include definite young-earth creationists like Nelson, deniers of
human ancestry like Meyer, Moonies like Wells, and so on. Face it: the
DI is a religious and political organization, not a scientific one.

> That's a tall order for a polemicist like you, so I'll leave you with
> this middle-school-level approximation to what I mean:
>
> Creationism : "proof" :: ID theory : evidence

I see that as nonsense. First, there is no "ID theory"; there is only
attacks on evolution. Second, I see no such distinction. "Creation
science" offers what it claims to be evidence just as much as IDers do.

> Even that is a tall order, because it's hard for you to resist the urge
> to pander to people who use "no evidence" to mean "no evidence sufficient
> to leave me at a loss for an answer."
>
> I do acknowledge that their evidence is meager except in the eyes of
> most believing Christians, just as your evidence for God's
> nonexistence is meager except in the eyes of your fellow militant atheists.

I reject your false equivalence.

gkaplan

unread,
Sep 30, 2016, 1:24:58 PM9/30/16
to talk-o...@moderators.isc.org
I am looking at a particular difference between humans and the primates
from which they putatively evolved. A paper that outlines this is: //
Large-scale sequencing of the CD33-related Siglec gene cluster in five
mammalian species reveals rapid evolution by multiple mechanisms
ww.pnas.org/cgi/doi/10.1073/pnas.0404833101 I also have two other papers.

The before and after includes three exons each with a number of
synonymous and non-synonymous differences. Before I get barraged with
criticisms that evolution does not have specific goals like this, let me
pre-emt this. I am merely using this as an example of the magnitude of
changes from a real world example and presuming that other things like
this are of the same order of magnitude.

I am thinking about a project that might get very involved, and which
may never be completed, to use scilab to step through the process using
mutation rates, etc from the peer reviewed papers. The skeleton of
this first pass at planning the procedure follows. I am very interested
in any comments/suggestions.

---

// Large-scale sequencing of the CD33-related Siglec gene cluster
// in five mammalian species reveals rapid evolution by multiple mechanisms
// www.pnas.org/cgi/doi/10.1073/pnas.0404833101 PNAS
// September 7, 2004 vol. 101 no. 36 13251–13256
// George Kaplan 9/29/16
// Scilab 5.5.2


// Exon Sd Nd S N pN pS
//
-----------------------------------------------------------------------------------
// Human–chimpanzee comparison
// Ig1 7 42 784.25 2431.75 1.935
// Ig2 17 19 582.92 1631.08 0.399
// Ig3 14 11 494.74 1332.26 0.292

// Glossary
// Ign, nth Ig-like domain of a Siglec;
// BAC, bacterial artificial chromosome;
// KLK, kallikrein;
// MCSs, multispecies conserved sequences;
// pS,synonymous substitution rate (SdS);
// Sia, sialic acid;
// Neu5Gc, N-glycolylneuraminic acid;
// CD33rSiglecs, CD33-related Siglecs.
Sd1 = 7 // Sd, no. of synonymous differences;
Sd2 = 17
Sd3 = 14
Nd1 = 42 // Nd, no. of nonsynonymous differences;
Nd2 = 19
Nd3 = 11
S1 = 784.25 // S, no. of synonymous sites;
S2 = 582.92
S3 = 494.74
Ns1 = 2431.75 // N, no. of nonsynonymous sites, Exon 1
Ns2 = 1631.08
Ns3 = 1332.26
pNpS1 = 1.935 // pN, nonsynonymous substitution rate (NdN);
pNpS2 = 0.399
pNpS3 = 0.292


// Initialize variables
mu = 0.000000011 // The human mutation rate per
nucleotide site per generation
t = 5000000 // Time since speciation
N = 10000 // Number of individuals in the population
Ne_N = 0.8 // Ne/N = 0.8 in man, Crow 1954
BP = 3000000000 // Number of base pairs in the genome.
gen = 20 // human generation length
HGC = 22500 // Human gene count

// Calculated
Ne = Ne_N * N // Variance effective number Kimura and
Crow 1963
Q = 4 * Ne * mu // Nucleotide diversity?
k = (2 * mu * t) / gen // Rate of accumulation of nucleotide.
sub] per gen
mu_gen = mu * BP * 2 // Mutations per generaion
mu_t = mu_gen * N * t // Mutations per population in time
p_site_mu = mu_t / BP // Probability of particular site
mutation in years
n_fix = 4 * Ne * gen // Time to fix neutral mutation

// Need
// Types and frequencies of different mutations
// gene copies
// deleterious mutations (insertions, deletions, etc)
// successful substitution of base

// Loop through generations
// Create gene copies at proper frequency.
// Loop through gene copies for this gene.
// Keep track of loss due to drift in pop for statistics

// Mutate organism based on number accumulated
// Randomly select mutation and fate of organism

// Mutate gene copy if randomly selected
// Randomly select mutation and fate of gene copy
// Calculate prob in gen of mut for each site in
// If Sprob >= 1, randomly mutate that site in array
// Randomly select ben. or del. mutation by site
// Update counter and status of site
// End Mutate gene copy if randomly selected

// End Mutate organism based on no. accumulated

// End Loop through gene copies for this gene
// End Create gene copies
// End loop through generations






Mark Isaak

unread,
Sep 30, 2016, 1:49:58 PM9/30/16
to talk-o...@moderators.isc.org
On 9/29/16 2:49 PM, John Harshman wrote:
> On 9/29/16 1:51 PM, Alan Kleinman MD PhD wrote:
>> On Thursday, September 29, 2016 at 12:45:01 PM UTC-7, John Harshman
>> wrote:
>>> On 9/29/16 12:30 PM, Alan Kleinman MD PhD wrote:
> [...]
> >>> Peter, explain to us why the evolution of a pseudogene would be any
>>>> different than the evolution of any other gene. The mathematics that
>>>> governs any evolutionary trajectory is simply a set of joint binomial
>>>> probability problems. And the way that natural selection improves the
>>>> probability for the next binomial probability problem in that set is
>>>> by amplification of the particular variant. Only if a mutation occurs
>>>> in that pseudogene that gives beneficial function improving that
>>>> members ability to replicate will help that variant on its particular
>>>> evolutionary trajectory.
>>>
>>> Pseudogenes (or at least the vast majority of them) don't have
>>> "beneficial function", hence no selection. That's why they're
>>> pseudogenes instead of genes. The only thing that would "help that
>>> variant" is chance, i.e. drift. Would you agree that drift can happen
>>> efficiently (as efficiently as it ever happens) in many loci at once?
>>
>> Drift occurs when selection is random. And random selection can
>> create population bottlenecks. But what is efficient in this case?
>
> Your use of non-standard language makes it difficult to communicate with
> you. Why do you use it? [...]

I like that he uses it. It tells everybody else that he is a crank,
with no real understanding of genetics and no intention ever of getting any.

--
Mark Isaak eciton (at) curioustaxonomy (dot) net
"We are not looking for answers. We are looking to come to an
understanding, recognizing that it is temporary--leaving us open to an
even richer understanding as further evidence surfaces." - author unknown

Alan Kleinman MD PhD

unread,
Sep 30, 2016, 2:05:02 PM9/30/16
to talk-o...@moderators.isc.org
A crank who published the correct physics and mathematics of rmns. And this is the mathematics and physics which describes how drug resistance occurs and why cancer treatments fail due to rmns. I like when you post videos which substantiate my mathematical model. Keep on cranking out the empirical evidence which supports my model.

Peter Nyikos

unread,
Sep 30, 2016, 3:54:58 PM9/30/16
to talk-o...@moderators.isc.org
On Friday, September 30, 2016 at 1:14:59 AM UTC-4, John Harshman wrote:
> On 9/29/16 6:52 PM, Peter Nyikos wrote:
> > On Thursday, September 29, 2016 at 2:10:01 PM UTC-4, John Harshman wrote:
> >> On 9/29/16 10:44 AM, Peter Nyikos wrote:
>
> >>> Do you think that particular pseudogene was not having any effects
> >>> at all for something like 50 million years? In that case, what's
> >>> the big deal about a deity putting it in?
> >>
> >> 50 million years turns out to be too long, probably. In fact it isn't
> >> clear, but the duplication could be as old as 30 million years or as
> >> recent as 8. And the pseudogenization almost as old as 8 or as recent as
> >> 5.8. Now, I would presume that the duplicate has been evolving neutrally
> >> ever since the duplication, since the original would compensate for any
> >> deactivating mutation in the copy.
> >
> > That last sentence is sheer speculation, unless there is something in
> > the linked article to support it. [I've only read the abstract, the
> > rest being paywalled; but I can get at it at my office tomorrow.]
>
> No, it's just how duplicate genes work. Since there is the original copy
> still serving the original function, the new copy is a spare, and
> deactivation can't be deleterious.

Wrong referent: I was questioning your claim that the duplicate had no effect
on the phenotype once it became a pseudogene.

> >> The pseudogene in humans is interesting in that recombination between
> >> the pseudogene and the functional copy is known to result in genetic
> >> disease in humans, possibly exacerbated by certain mutations in the
> >> pseudogene. So there's an example of a deleterious mutation happening in
> >> a functionless sequence. I would expect that there is now some selection
> >> for deletion, i.e. if another big deletion happened in the pseudogene,
> >> recombination would be less likely to occur.
> >>
> >> Why would a deity put in a useless, or at best a pointless extra, copy
> >> of a functional gene?
> >
> > Unfortunately, George shows no inclination to look for
> > information about the usefulness of some pseudogenes, and I don't have
> > the time for it, for another month at least. So we'll have to be
> > stalemated about the alleged uselessness of this particular pseudogene.
>
> No we won't. This particular pseudogene is useless.

More speculation by you.

> At the very least it
> should be assumed useless until there's good evidence otherwise, as most
> pseudogenes are useless.

Determining whether a pseudogene is useless is an arduous process,
since it involves influence on the expression of other genes.
You have no reason to believe that the tiny handful for which
some use has been determined is anything like the true number.

To paraphrase Bill Rogers: who is going to try to carry out the
requisite experiments? There is no drive for grant funding of them.

> And in fact your test has more or less been
> performed. It's evolving much faster than conserved sequences.

That's what I meant by the following test. Will you claim that
I was unclear about what I meant, as per your usual custom?

> >>> One way to test this hypothesis is to see how many alleles that
> >>> one pseudogene has accumulated. At minimum, I would expect
> >>> every single primate species to have alleles that are not found in
> >>> any other primate species.
> >>
> >> What hypothesis? The one about neutral evolution for 50 million years or
> >> the one about a deity?

> > Don't be silly: of course I meant the former, except that now we see
> > that it is a much shorter time, and we only have four or five species
> > to deal with.
> >
> > Wait, make that six. By population genetic standards, the mountain
> > gorilla is a separate species from the lowland gorilla.
>
> No, still 5. There is sequence only for one of those, and in fact one
> individual.

I meant future dealings, so to speak, to include the bonobo as well.
As per the test I was proposing.


<snip comments to be dealt with in separate reply>


> >> I don't think you're talking about allelic
> >> variation here, really, but about divergence among species.
> >
> > The two go hand in hand.
>
> They do only if certain assumptions about population size stability are
> met, which they are not.

You are going off on a tangent here. The test I proposed darn well
makes use of allelic variation, and you have belatedly confirmed
that it passes the test.

> Humans, you have been told, have less allelic
> diversity than most species, including chimps.

Irrelevant to my proposed test, which has to do with ONE (pseudo)gene and
chimps, bonobos, mountain and lowland gorillas, and orangutans.

By the way, do you include Neanderthals and Denisovians among "humans"?

> >> You mean to
> >> consider whether the pseudogene has accumulated changes at the rate
> >> expected from neutral evolution.
> >
> > Totally neutral, as befits a totally non-functioning pseudogene.
> > [And I don't mean just non-protein-coding, which is what all too many
> > talk.origins participants think "non-functional" means.]
>
> Do you know of anyone who thinks that?

Google Discovery Institute. They use the word "function" in
the natural way, not the Humpty Dumpty way you and Vince favor.

Remainder deleted, to be replied to later.

Peter Nyikos
Professor, Department of Math. -- standard disclaimer --
U. of South Carolina, Columbia
http://www.math.sc.edu/~nyikos/

John Harshman

unread,
Sep 30, 2016, 4:39:58 PM9/30/16
to talk-o...@moderators.isc.org
On 9/30/16 12:53 PM, Peter Nyikos wrote:
> On Friday, September 30, 2016 at 1:14:59 AM UTC-4, John Harshman wrote:
>> On 9/29/16 6:52 PM, Peter Nyikos wrote:
>>> On Thursday, September 29, 2016 at 2:10:01 PM UTC-4, John Harshman wrote:
>>>> On 9/29/16 10:44 AM, Peter Nyikos wrote:
>>
>>>>> Do you think that particular pseudogene was not having any effects
>>>>> at all for something like 50 million years? In that case, what's
>>>>> the big deal about a deity putting it in?
>>>>
>>>> 50 million years turns out to be too long, probably. In fact it isn't
>>>> clear, but the duplication could be as old as 30 million years or as
>>>> recent as 8. And the pseudogenization almost as old as 8 or as recent as
>>>> 5.8. Now, I would presume that the duplicate has been evolving neutrally
>>>> ever since the duplication, since the original would compensate for any
>>>> deactivating mutation in the copy.
>>>
>>> That last sentence is sheer speculation, unless there is something in
>>> the linked article to support it. [I've only read the abstract, the
>>> rest being paywalled; but I can get at it at my office tomorrow.]
>>
>> No, it's just how duplicate genes work. Since there is the original copy
>> still serving the original function, the new copy is a spare, and
>> deactivation can't be deleterious.
>
> Wrong referent: I was questioning your claim that the duplicate had no effect
> on the phenotype once it became a pseudogene.

I don't recall making such a claim, though it's certainly the way to
bet. Genes affect the phenotype by being expressed. Pseudogenes are not
expressed. They can't have effects on the phenotype unless they evolve
new functions.

>>>> The pseudogene in humans is interesting in that recombination between
>>>> the pseudogene and the functional copy is known to result in genetic
>>>> disease in humans, possibly exacerbated by certain mutations in the
>>>> pseudogene. So there's an example of a deleterious mutation happening in
>>>> a functionless sequence. I would expect that there is now some selection
>>>> for deletion, i.e. if another big deletion happened in the pseudogene,
>>>> recombination would be less likely to occur.
>>>>
>>>> Why would a deity put in a useless, or at best a pointless extra, copy
>>>> of a functional gene?
>>>
>>> Unfortunately, George shows no inclination to look for
>>> information about the usefulness of some pseudogenes, and I don't have
>>> the time for it, for another month at least. So we'll have to be
>>> stalemated about the alleged uselessness of this particular pseudogene.
>>
>> No we won't. This particular pseudogene is useless.
>
> More speculation by you.

The abstract of the paper I cited suggests it is. I see no evidence that
it isn't.

>> At the very least it
>> should be assumed useless until there's good evidence otherwise, as most
>> pseudogenes are useless.
>
> Determining whether a pseudogene is useless is an arduous process,
> since it involves influence on the expression of other genes.
> You have no reason to believe that the tiny handful for which
> some use has been determined is anything like the true number.

I'm actually talking about the test you yourself proposed: sequence
conservation. That may not be a perfect test, but it should certainly
work in the great majority of cases. You don't agree?

> To paraphrase Bill Rogers: who is going to try to carry out the
> requisite experiments? There is no drive for grant funding of them.

Are you just being a contrarian here, or do you have some reason to
suppose that pseudogenes are not overwhelmingly useless?

>> And in fact your test has more or less been
>> performed. It's evolving much faster than conserved sequences.
>
> That's what I meant by the following test. Will you claim that
> I was unclear about what I meant, as per your usual custom?

You were unclear, but I knew what you were trying to say, as I
explained. So, since the test has been performed, can we agree that it's
useless?

>>>>> One way to test this hypothesis is to see how many alleles that
>>>>> one pseudogene has accumulated. At minimum, I would expect
>>>>> every single primate species to have alleles that are not found in
>>>>> any other primate species.
>>>>
>>>> What hypothesis? The one about neutral evolution for 50 million years or
>>>> the one about a deity?
>
>>> Don't be silly: of course I meant the former, except that now we see
>>> that it is a much shorter time, and we only have four or five species
>>> to deal with.
>>>
>>> Wait, make that six. By population genetic standards, the mountain
>>> gorilla is a separate species from the lowland gorilla.
>>
>> No, still 5. There is sequence only for one of those, and in fact one
>> individual.
>
> I meant future dealings, so to speak, to include the bonobo as well.
> As per the test I was proposing.

Why isn't the current sample good enough for the test?

> <snip comments to be dealt with in separate reply>
>
>
>>>> I don't think you're talking about allelic
>>>> variation here, really, but about divergence among species.
>>>
>>> The two go hand in hand.
>>
>> They do only if certain assumptions about population size stability are
>> met, which they are not.
>
> You are going off on a tangent here. The test I proposed darn well
> makes use of allelic variation, and you have belatedly confirmed
> that it passes the test.

You may not understand the difference between allelic variation (within
a population) and divergence (between populations). They are often
correlated, but they aren't the same thing.

>> Humans, you have been told, have less allelic
>> diversity than most species, including chimps.
>
> Irrelevant to my proposed test, which has to do with ONE (pseudo)gene and
> chimps, bonobos, mountain and lowland gorillas, and orangutans.

I don't think you know what you're proposing. You keep saying "allelic
variation" when you mean "divergence". I'm talking about allelic
diversity at any single locus, which is in fact what the term means.

> By the way, do you include Neanderthals and Denisovians among "humans"?

I generally do, though the word is ambiguous. Why?

>>>> You mean to
>>>> consider whether the pseudogene has accumulated changes at the rate
>>>> expected from neutral evolution.
>>>
>>> Totally neutral, as befits a totally non-functioning pseudogene.
>>> [And I don't mean just non-protein-coding, which is what all too many
>>> talk.origins participants think "non-functional" means.]
>>
>> Do you know of anyone who thinks that?
>
> Google Discovery Institute. They use the word "function" in
> the natural way, not the Humpty Dumpty way you and Vince favor.

That doesn't seem to have answered my question. And I have no idea what
you think is odd about the definition of "function" I favor, or even
what definition Vince likes, or you like. Why the DI is relevant here
also escapes me. So who thinks that "non-protein-coding" is the
definition of "non-functional"?

Peter Nyikos

unread,
Sep 30, 2016, 6:39:57 PM9/30/16
to talk-o...@moderators.isc.org
On Friday, September 30, 2016 at 4:39:58 PM UTC-4, John Harshman wrote:
> On 9/30/16 12:53 PM, Peter Nyikos wrote:
> > On Friday, September 30, 2016 at 1:14:59 AM UTC-4, John Harshman wrote:
> >> On 9/29/16 6:52 PM, Peter Nyikos wrote:
> >>> On Thursday, September 29, 2016 at 2:10:01 PM UTC-4, John Harshman wrote:
> >>>> On 9/29/16 10:44 AM, Peter Nyikos wrote:
> >>
> >>>>> Do you think that particular pseudogene was not having any effects
> >>>>> at all for something like 50 million years? In that case, what's
> >>>>> the big deal about a deity putting it in?
> >>>>
> >>>> 50 million years turns out to be too long, probably. In fact it isn't
> >>>> clear, but the duplication could be as old as 30 million years or as
> >>>> recent as 8. And the pseudogenization almost as old as 8 or as recent as
> >>>> 5.8. Now, I would presume that the duplicate has been evolving neutrally
> >>>> ever since the duplication, since the original would compensate for any
> >>>> deactivating mutation in the copy.
> >>>
> >>> That last sentence is sheer speculation, unless there is something in
> >>> the linked article to support it. [I've only read the abstract, the
> >>> rest being paywalled; but I can get at it at my office tomorrow.]
> >>
> >> No, it's just how duplicate genes work. Since there is the original copy
> >> still serving the original function, the new copy is a spare, and
> >> deactivation can't be deleterious.
> >
> > Wrong referent: I was questioning your claim that the duplicate had no effect
> > on the phenotype once it became a pseudogene.
>
> I don't recall making such a claim,

[taken directly from the sentence that I was labeling "speculation" above:]
Now, I would presume that the duplicate has been evolving neutrally
ever since the duplication

...which of course includes the stretch of time after it became
a pseudogene, which evidently means "deactivation" to you.

> though it's certainly the way to
> bet. Genes affect the phenotype by being expressed. Pseudogenes are not
> expressed. They can't have effects on the phenotype unless they evolve
> new functions.

You really are hedging your bets with that last sentence. That's
welcome progress.

> >>>> The pseudogene in humans is interesting in that recombination between
> >>>> the pseudogene and the functional copy is known to result in genetic
> >>>> disease in humans, possibly exacerbated by certain mutations in the
> >>>> pseudogene. So there's an example of a deleterious mutation happening in
> >>>> a functionless sequence. I would expect that there is now some selection
> >>>> for deletion, i.e. if another big deletion happened in the pseudogene,
> >>>> recombination would be less likely to occur.
> >>>>
> >>>> Why would a deity put in a useless, or at best a pointless extra, copy
> >>>> of a functional gene?
> >>>
> >>> Unfortunately, George shows no inclination to look for
> >>> information about the usefulness of some pseudogenes, and I don't have
> >>> the time for it, for another month at least. So we'll have to be
> >>> stalemated about the alleged uselessness of this particular pseudogene.
> >>
> >> No we won't. This particular pseudogene is useless.
> >
> > More speculation by you.
>
> The abstract of the paper I cited suggests it is.

There is no "suggests" about it. It says the copy with 55 deleted is
"non-functional," and so it either is saying that

(1) it has no protein coding function, in which case it is definitely
not suggesting it is useless or

(2) it really does mean having no effect whatsoever on the phenotype,
not even indirectly by its position causing modification of the
expression of some genes.

<snip of things to be dealt with in separate reply>

> >>>> You mean to
> >>>> consider whether the pseudogene has accumulated changes at the rate
> >>>> expected from neutral evolution.
> >>>
> >>> Totally neutral, as befits a totally non-functioning pseudogene.
> >>> [And I don't mean just non-protein-coding, which is what all too many
> >>> talk.origins participants think "non-functional" means.]
> >>
> >> Do you know of anyone who thinks that?
> >
> > Google Discovery Institute. They use the word "function" in
> > the natural way, not the Humpty Dumpty way you and Vince favor.
>
> That doesn't seem to have answered my question. And I have no idea what
> you think is odd about the definition of "function" I favor, or even
> what definition Vince likes, or you like. Why the DI is relevant here
> also escapes me. So who thinks that "non-protein-coding" is the
> definition of "non-functional"?

Vince pretended to, by emphasizing "to no effect":

> Why would your Genetic Engineer (capitalized to reflect who I think
> you think this genetic engineer is) choose 55 base pairs to delete,
> *to no effect*?

When I called him to task for this, I made the tactical error of
not interrupting here [a rarity for me] and waiting until he
also added the following:

> (Remember that pseudogenes have no function related to their sequence
> -- John mentioned this as well, but you seem to have drifted off from
> what he was telling you.)

Vince took advantage of that in reply to my taking him to task. He
snipped the first of the two sentences by him that you see above,
and playing innocent.

You also favored it in the minds of casual readers by spin-doctoring
Vince's claim (*to no effect*) into a sneaky piece of propaganda, saying
that "to a first approximation" pseudogenes have no effect.

As has often happened throughout these last close-to-6 years, the issue
is not what you privately think but the persona that you exhibit
in this newsgroup.

Peter Nyikos
Professor, Dept. of Mathematics -- standard disclaimer --
U. of S. Carolina at Columbia
http://people.math.sc.edu/nyikos

John Harshman

unread,
Sep 30, 2016, 7:29:57 PM9/30/16
to talk-o...@moderators.isc.org
Right. So what you said isn't what I said. I don't say it had no effect
on phenotype after becoming a pseudogene. I say it had no effect after
being duplicated. It might be that it gained an effect later, but I have
no reason to believe it and every reason to doubt it.

>> though it's certainly the way to
>> bet. Genes affect the phenotype by being expressed. Pseudogenes are not
>> expressed. They can't have effects on the phenotype unless they evolve
>> new functions.
>
> You really are hedging your bets with that last sentence. That's
> welcome progress.

Do you think this new function is likely?

>>>>>> The pseudogene in humans is interesting in that recombination between
>>>>>> the pseudogene and the functional copy is known to result in genetic
>>>>>> disease in humans, possibly exacerbated by certain mutations in the
>>>>>> pseudogene. So there's an example of a deleterious mutation happening in
>>>>>> a functionless sequence. I would expect that there is now some selection
>>>>>> for deletion, i.e. if another big deletion happened in the pseudogene,
>>>>>> recombination would be less likely to occur.
>>>>>>
>>>>>> Why would a deity put in a useless, or at best a pointless extra, copy
>>>>>> of a functional gene?
>>>>>
>>>>> Unfortunately, George shows no inclination to look for
>>>>> information about the usefulness of some pseudogenes, and I don't have
>>>>> the time for it, for another month at least. So we'll have to be
>>>>> stalemated about the alleged uselessness of this particular pseudogene.
>>>>
>>>> No we won't. This particular pseudogene is useless.
>>>
>>> More speculation by you.
>>
>> The abstract of the paper I cited suggests it is.
>
> There is no "suggests" about it. It says the copy with 55 deleted is
> "non-functional," and so it either is saying that
>
> (1) it has no protein coding function, in which case it is definitely
> not suggesting it is useless or
>
> (2) it really does mean having no effect whatsoever on the phenotype,
> not even indirectly by its position causing modification of the
> expression of some genes.

I suggest that "non-functional" means "non-functional". It has no
function. I suppose it could conceivably have an effect on phenotype if
that effect were itself not subject to selection, i.e. neutral. But this
also is not the way to bet. You are clinging to seriously remote
possibilities as if they're probable. Why?

Also, I would not equate "function" with "effect on phenotype". A
sequence that causes cancer has an effect on phenotype, but I wouldn't
say it has a function.

>>>>>> You mean to
>>>>>> consider whether the pseudogene has accumulated changes at the rate
>>>>>> expected from neutral evolution.
>>>>>
>>>>> Totally neutral, as befits a totally non-functioning pseudogene.
>>>>> [And I don't mean just non-protein-coding, which is what all too many
>>>>> talk.origins participants think "non-functional" means.]
>>>>
>>>> Do you know of anyone who thinks that?
>>>
>>> Google Discovery Institute. They use the word "function" in
>>> the natural way, not the Humpty Dumpty way you and Vince favor.
>>
>> That doesn't seem to have answered my question. And I have no idea what
>> you think is odd about the definition of "function" I favor, or even
>> what definition Vince likes, or you like. Why the DI is relevant here
>> also escapes me. So who thinks that "non-protein-coding" is the
>> definition of "non-functional"?
>
> Vince pretended to, by emphasizing "to no effect":

In one particular case, for which the evidence is good. I see no reason
to suppose he meant it as a general statement.

>> Why would your Genetic Engineer (capitalized to reflect who I think
>> you think this genetic engineer is) choose 55 base pairs to delete,
>> *to no effect*?
>
> When I called him to task for this, I made the tactical error of
> not interrupting here [a rarity for me] and waiting until he
> also added the following:
>
>> (Remember that pseudogenes have no function related to their sequence
>> -- John mentioned this as well, but you seem to have drifted off from
>> what he was telling you.)
>
> Vince took advantage of that in reply to my taking him to task. He
> snipped the first of the two sentences by him that you see above,
> and playing innocent.
>
> You also favored it in the minds of casual readers by spin-doctoring
> Vince's claim (*to no effect*) into a sneaky piece of propaganda, saying
> that "to a first approximation" pseudogenes have no effect.

The earth is spherical. Agree or disagree? All scientific statements are
approximate. (Or should I say that approximately all are approximate?)
The few pseudogenes that have functions can be ignored for most
purposes. Though I think Vince is wrong about one thing: the few
functional pseudogenes do seem to have functions that depend on
sequence. Or he may be talking about just this one pseudogene, for which
there is evidence of no function that depends on sequence; there could
still be a function that doesn't, but that doesn't seem likely.

> As has often happened throughout these last close-to-6 years, the issue
> is not what you privately think but the persona that you exhibit
> in this newsgroup.

Your paranoia is once more in evidence.


Alan Kleinman MD PhD

unread,
Sep 30, 2016, 8:04:58 PM9/30/16
to talk-o...@moderators.isc.org
It's a strange way evolutionists use the English language, using a word which means merge to describe a phenomenon which is diverging. From an engineering point of view, drift can create bottlenecks in the population. But think about this, when do directional selection pressures become non-directional? And why do you include links which discuss allele frequencies when this has nothing to do with rmns?

When are you going to get your buddies at Harvard to run their family size agar plate with 2 and 3 drugs? Don't you want to prove me wrong?

Bill Rogers

unread,
Sep 30, 2016, 8:59:57 PM9/30/16
to talk-o...@moderators.isc.org
Well, since you are an expert on the mathematics of evolution and population genetics I was sure you'd be familiar with coalescent theory; but since you seemed to be surprised at how the word is used on the field, I thought I'd give you some links for background. Surely the math isn't particularly hard for you.

As for proving you wrong, I'm not interested in that in the least.

John Harshman

unread,
Sep 30, 2016, 9:04:57 PM9/30/16
to talk-o...@moderators.isc.org
Nevertheless, it's the standard term, which you should conform to if you
want to make sense. The reason, of course, is that geneticists often
start in the present and infer the past. Why, many of them even draw
their trees upside down, with the root on top.

> From an engineering point of view, drift can create bottlenecks in
> the population.

I don't know what an engineering point of view has to do with it, or how
it would create a bottleneck, unless this is your personal definition of
"bottleneck", meaning coalescence.

> But think about this, when do directional selection
> pressures become non-directional? And why do you include links which
> discuss allele frequencies when this has nothing to do with rmns?

!

Alan Kleinman MD PhD

unread,
Sep 30, 2016, 9:29:57 PM9/30/16
to talk-o...@moderators.isc.org
Since you are an expert in Malaria drug resistance, I thought you would be familiar with the mathematics of rmns. I've never claimed I know every theory evolutionists float out there. But I do claim that I understand the mathematics of Haldane and Kimura and I know where they made the error in their physics. However, I do claim that I got the physics and mathematics of rmns correct. And the evolutionist view of coalescence is a totally useless theory for understanding evolution by rmns. Do you still think relative frequencies of alleles has anything to do with the emergence of drug resistance with Malaria?
>
> As for proving you wrong, I'm not interested in that in the least.
There's one thing you don't have to worry about, you won't.


Alan Kleinman MD PhD

unread,
Sep 30, 2016, 9:54:57 PM9/30/16
to talk-o...@moderators.isc.org
Make sense???? Evolutionism has nothing to do with making sense, it's a belief system based on irrational extrapolations and speculations.
>
> > From an engineering point of view, drift can create bottlenecks in
> > the population.
>
> I don't know what an engineering point of view has to do with it, or how
> it would create a bottleneck, unless this is your personal definition of
> "bottleneck", meaning coalescence.
What the bottleneck does is pass all the genetics of that progenitor to all descendants. Every allele from the progenitor is fixed in all the descendants aside from any mutations accumulated over generations in the descendants.
>
> > But think about this, when do directional selection
> > pressures become non-directional? And why do you include links which
> > discuss allele frequencies when this has nothing to do with rmns?
>
> !
Having a hard time answering the questions? You are the one who pointed me to the answer to this question. Let me rephrase the question using an example. Single drug therapy for HIV is a directional selection pressure, two drug therapy for HIV are directional selection pressures, does three drug therapy consist of directional selection pressures? If they are directional selection pressures, why doesn't HIV accumulate the beneficial mutations? If they are not directional selection pressures, then how is the remaining population evolving? Do you think they are coalescing?

John Harshman

unread,
Sep 30, 2016, 10:04:57 PM9/30/16
to talk-o...@moderators.isc.org
Again, that isn't a bottleneck. It's a coalescent. And this genomic
coalescent is true only for a clonal population.

>>> But think about this, when do directional selection
>>> pressures become non-directional? And why do you include links which
>>> discuss allele frequencies when this has nothing to do with rmns?
>>
>> !

> Having a hard time answering the questions? You are the one who
> pointed me to the answer to this question. Let me rephrase the
> question using an example. Single drug therapy for HIV is a
> directional selection pressure, two drug therapy for HIV are
> directional selection pressures, does three drug therapy consist of
> directional selection pressures? If they are directional selection
> pressures, why doesn't HIV accumulate the beneficial mutations? If
> they are not directional selection pressures, then how is the
> remaining population evolving? Do you think they are coalescing?

Once more, in the scenarios you are talking about, there is no
directional selection unless an organism has resistance to all the
"selection pressures". In order for selection to occur there must be
genetic variation, some of which has higher fitness than others, and
there's none of that in your scenarios. Most situations don't work that
way. But yes, the populations are coalescing if they are reproducing at
all, purely by drift.

Peter Nyikos

unread,
Sep 30, 2016, 10:44:57 PM9/30/16
to talk-o...@moderators.isc.org

On Friday, September 30, 2016 at 4:39:58 PM UTC-4, John Harshman wrote:
> On 9/30/16 12:53 PM, Peter Nyikos wrote:

> > [For that matter, those standards probably give us at least a dozen
> > different species [= reproductively isolated tribes] of human beings on
> > various hilltops in New Guinea, but that issue would call for a
separate
> > thread.]
>
> It's a silly issue. There are no isolated tribes.

The fact that you left off "reproductively" is very telling.

Did you know that there are over 800 different languages just in
Papua New Guinea? and that quite a few of them are "isolates"
with no known related languages? [Basque and Ainu are the two best known
of the many language isolates of the world.] See:

http://www.ethnologue.com/country/PG
maps:
http://www.ethnologue.com/country/PG/maps
http://www.ethnologue.com/map/PG_xx [index map]

All this is highly suggestive of centuries of reproductive isolation,
of the relative sort that population geneticists tolerate. [Some
infrequent interbreeding, but not enough for the hybrids to become
established -- isn't that the way you put it on another thread?]

<snip of much text -- some dealt with, some to be dealt with>

> > Are you suggesting that you read everything on that "parent" thread?
>
> I read enough to know what's being talked about.
>
> > If so, I wonder whether you'll deign to address any of what I've
> > written to Kleinman these last two days about Sphenosuchia
> > and other evolutionary matters.
>
> No, I'm just going to watch with amusement for a while while you fight
> it out.

That decision went up in smoke as you got busy on my new thread,
"Bird Origins Again." But if I hadn't started it, your continued
absence would have been evidence that I am more interested in the
scientific issues than you are.

And you might still provide evidence for that on the new thread. Time will
tell.

> Remember, Kleinman denies that most evolution can actually
> happen. He doesn't care whether birds are not dinosaurs or not
> thecodonts. They're not descended from anything but were created as
birds.

Maybe he is as closed minded as you say. But he may yet be amenable
to a fragment of bird evolution roughly the size of the horse family
tree. That's something you are ill equipped to deal with, due
to your unreasoning prejudice against paraphyletic taxa.

Peter Nyikos
Professor, Dept. of Mathematics -- standard disclaimer --
U. of S. Carolina at Cola., SC
http://people.math.sc.edu/nyikos

Alan Kleinman MD PhD

unread,
Sep 30, 2016, 10:59:58 PM9/30/16
to talk-o...@moderators.isc.org
Whatever word you want to use, it's subpopulation that has arose from a single progenitor (in a clonal population example).
>
> >>> But think about this, when do directional selection
> >>> pressures become non-directional? And why do you include links which
> >>> discuss allele frequencies when this has nothing to do with rmns?
> >>
> >> !
>
> > Having a hard time answering the questions? You are the one who
> > pointed me to the answer to this question. Let me rephrase the
> > question using an example. Single drug therapy for HIV is a
> > directional selection pressure, two drug therapy for HIV are
> > directional selection pressures, does three drug therapy consist of
> > directional selection pressures? If they are directional selection
> > pressures, why doesn't HIV accumulate the beneficial mutations? If
> > they are not directional selection pressures, then how is the
> > remaining population evolving? Do you think they are coalescing?
>
> Once more, in the scenarios you are talking about, there is no
> directional selection unless an organism has resistance to all the
> "selection pressures". In order for selection to occur there must be
> genetic variation, some of which has higher fitness than others, and
> there's none of that in your scenarios. Most situations don't work that
> way. But yes, the populations are coalescing if they are reproducing at
> all, purely by drift.
That's right, when a population is subjected to multiple selection pressures, for which it can't get a single beneficial mutation and the population is not driven to extinction, the population can only drift. So give us some examples of scenarios that don't work the way the scenarios I just described.

Alan Kleinman MD PhD

unread,
Sep 30, 2016, 11:04:57 PM9/30/16
to talk-o...@moderators.isc.org
I'd have no problem with that. It's that reptile to bird thing that John is obsessed with which defies mathematical logic that I have a problem with.

John Harshman

unread,
Sep 30, 2016, 11:34:57 PM9/30/16
to talk-o...@moderators.isc.org
On 9/30/16 7:43 PM, Peter Nyikos wrote:
>
> On Friday, September 30, 2016 at 4:39:58 PM UTC-4, John Harshman wrote:
>> On 9/30/16 12:53 PM, Peter Nyikos wrote:
>
>>> [For that matter, those standards probably give us at least a dozen
> > > different species [= reproductively isolated tribes] of human beings on
> > > various hilltops in New Guinea, but that issue would call for a
> separate
> > > thread.]
> >
> > It's a silly issue. There are no isolated tribes.
>
> The fact that you left off "reproductively" is very telling.

No it isn't. You persist in inferring huge points from tiny differences
in phrasing.

> Did you know that there are over 800 different languages just in
> Papua New Guinea? and that quite a few of them are "isolates"
> with no known related languages? [Basque and Ainu are the two best known
> of the many language isolates of the world.] See:

Yes.

> http://www.ethnologue.com/country/PG
> maps:
> http://www.ethnologue.com/country/PG/maps
> http://www.ethnologue.com/map/PG_xx [index map]
>
> All this is highly suggestive of centuries of reproductive isolation,
> of the relative sort that population geneticists tolerate. [Some
> infrequent interbreeding, but not enough for the hybrids to become
> established -- isn't that the way you put it on another thread?]

No it isn't, and no I didn't. Bet there's very little genetic
differentiation between adjacent populations. How did we even get on
this subject?

> <snip of much text -- some dealt with, some to be dealt with>
>
> > > Are you suggesting that you read everything on that "parent" thread?
> >
> > I read enough to know what's being talked about.
> >
> > > If so, I wonder whether you'll deign to address any of what I've
> > > written to Kleinman these last two days about Sphenosuchia
> > > and other evolutionary matters.
> >
> > No, I'm just going to watch with amusement for a while while you fight
> > it out.
>
> That decision went up in smoke as you got busy on my new thread,
> "Bird Origins Again." But if I hadn't started it, your continued
> absence would have been evidence that I am more interested in the
> scientific issues than you are.

This is starting to become painful again. I might not keep responding
unless you tone down the crap.

> And you might still provide evidence for that on the new thread. Time will
> tell.
>
>> Remember, Kleinman denies that most evolution can actually
> > happen. He doesn't care whether birds are not dinosaurs or not
> > thecodonts. They're not descended from anything but were created as
> birds.
>
> Maybe he is as closed minded as you say. But he may yet be amenable
> to a fragment of bird evolution roughly the size of the horse family
> tree. That's something you are ill equipped to deal with, due
> to your unreasoning prejudice against paraphyletic taxa.

Go for it, since you're so well equipped.

John Harshman

unread,
Sep 30, 2016, 11:44:57 PM9/30/16
to talk-o...@moderators.isc.org
Correct. That isn't a bottleneck. And yes, it's only in a clonal population.
Any scenario in which single mutations are beneficial, i.e. are subject
to natural selection individually. Suppose we have two genes, each
responding to a different stress. Say each has a "good" allele and a
"bad" one. Call the good ones A and B, the bad ones a and b. And for
simplicity, let's call them haploid. Call genotype AB the standard, with
a fitness of 1. Ab = .7, aB = .8, and ab = .5. And let's give them a
temporary diploid phase in which two individuals join, recombine, and
separate. If we start out with single A and B mutants in separate
individuals, I suggest that both will "amplify". And eventually, there
will be AB individuals resulting from conjugation. What's wrong with
that scenario?


Alan Kleinman MD PhD

unread,
Oct 1, 2016, 12:49:58 AM10/1/16
to talk-o...@moderators.isc.org
There's nothing wrong with that scenario, in fact, it sounds something like HIV when only two drugs are used. But I don't think this is a typical evolutionary scenario. I don't believe that targeted selection pressures are typical. For example, the Lenski experiment starvation selection pressure targets multiple genetic loci. Thermal stress affects many biological molecules and would target multiple genetic loci.

And one more selection pressure targeting one of the genetic loci shuts down the amplification process. So how do you extrapolate a scenario like the one you propose for the transformation of a reptile population into a bird population? What are the selection pressures, what are the targeted genetic loci?

John Harshman

unread,
Oct 1, 2016, 1:39:56 AM10/1/16
to talk-o...@moderators.isc.org
No, it's nothing like HIV when only two drugs are used. Under your
assumption, resistance to one drug alone is not beneficial. Under your
scenario, the fitness would be something like Ab=aB=ab = .5, AB = 1. See
the difference?

> And one more selection pressure targeting one of the genetic loci
> shuts down the amplification process.

Why?

> So how do you extrapolate a
> scenario like the one you propose for the transformation of a reptile
> population into a bird population? What are the selection pressures,
> what are the targeted genetic loci?

Why this obsession with reptiles?

Bill Rogers

unread,
Oct 1, 2016, 8:04:59 AM10/1/16
to talk-o...@moderators.isc.org
On Friday, September 30, 2016 at 9:29:57 PM UTC-4, Alan Kleinman MD PhD wrote:

> > As for proving you wrong, I'm not interested in that in the least.
> There's one thing you don't have to worry about, you won't.

I am quite sure you will never know what it feels like to be proven wrong.

Alan Kleinman MD PhD

unread,
Oct 1, 2016, 12:09:55 PM10/1/16
to talk-o...@moderators.isc.org
Feel free to quote me if I said that. What I said was that one drug therapy gives a directional selection pressure when treating HIV, usind two drug therapy give two directional selection pressures (that is a more complex evolutionary trajectory), however three drug therapy gives too many selection pressures (and too complex of an evolutionary trajectory) for the population to evolve to and thus leads to drift (and possibly coalescence) for that remaining population. For one and two drug therapy, the population can evolve resistance to the selection pressures for several reasons, population sizes, mutation rates and possibly recombination. I am making no assumption on the fitness for any genotype, I'm simply correlating your theoretical example to a real, measurable and repeatable example. You do understand that one and two drug therapy does not give durable treatment for HIV but three drug combinations do?
>
> > And one more selection pressure targeting one of the genetic loci
> > shuts down the amplification process.
>
> Why?
Because of the multiplication rule of probabilities. The evolutionary trajectory has become too complex and requires too many simultaneous mutations to improve fitness. The probabilities for this to happen drops to virtually nil.
>
> > So how do you extrapolate a
> > scenario like the one you propose for the transformation of a reptile
> > population into a bird population? What are the selection pressures,
> > what are the targeted genetic loci?
>
> Why this obsession with reptiles?

Because they are so cute and look so much like birds, except they have no feathers.

Alan Kleinman MD PhD

unread,
Oct 1, 2016, 12:19:55 PM10/1/16
to talk-o...@moderators.isc.org
Like you when I proved to you that the size of your study affects the probability of emergence of drug resistance? And you said, "who cares". I feel sorry for the people who suffer from Malaria when they have researchers like you wasting tax payers money trying to solve their problem.

RonO

unread,
Oct 1, 2016, 12:59:55 PM10/1/16
to talk-o...@moderators.isc.org
post 2

Your links do not work for some reason. This is the paper that you are
talking about.

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC516556/

These are not pseudogenes. There are a group of 7 genes that evolved by
gene duplication that cluster in one region of a chromosome. This just
means that there are 7 functional genes that evolved from one progenitor
genes and they are close to each other on a chromosome. This paper is
talking about a gene family (the cluster of 7 genes is a gene family)
that diverged from each other by positive selection. This just means
that this is not neutral evolution for some of the mutations that they
are talking about that occurred between the 7 genes. Not only that but
multiple types of mutations are discussed and the different types of
mutations have different rates of occurring. Any mutation rate argument
is not going to work because you have to figure out what the selection
intensity was and you have to separate the different types of mutations
under discussion.

You would not be able to figure out mutation rates from the sequence
itself. What you are observing are the fixed mutations between species,
and fixation in this case could be due to positive selection, and
different rates of mutation and so is not dependent on some general
mutation rate like single nucleotide neutral mutations are.

You would be just spinning your wheels and getting no where.

Pick a pseudo gene that was non functional and you would do better.

There are several types of psuedogenes. One type is a gene that was
once functional, but suffered a mutation that knocked out the gene
function. The vitamin C gene (Gulonolactone oxidase, GULO) is an
example of such a gene. Our primate ancestors up to prosimian
incectovores like Tarsiers have functional GULO genes, but part of the
gene looks like it got deleted in Monkeys on up to us. So we have a non
functional GULO gene that we share with the other apes and monkeys.

Another type of psuedogene is one where an RNA transcript is reinserted
back into the genome. It was an RNA sequence that may have been
processed (introns removed) and the RNA transcript was put back into the
DNA genome. These psuedogenes are not usually functional in that they
do not produce a functional protein product. They lack parts of the
gene that are required for expression and sometimes they are only
partially processed (only some of the introns are removed) or they may
be missing part of the coding sequence of the gene. There are
hemoglobin psuedo genes like this in humans.

The most common type of pseudogene is due to transposable elements.
These are parasitic DNA sequences that can replicate themselves and move
around in the genome. You have over 600,000 ALU transposable elements
in your genome comprising around 10% of your genome. So they can make a
lot of copies of themselves and pepper the genome with these copies.
Mutations occur in transposable elements that make them nonfunctional so
they don't make functional copies of themselves. They get stuck where
they are in the genome and their DNA sequences slowly degrade by
mutation until they can't be determined to have been transposable
elements any longer. Their sequence changes so much after millions of
years so that they look like random sequence. A significant fraction of
the 600,000 ALU sequences in your genome are pseudogenes degenerating
into random sequence at this time. There are other transposable
elements that include retrovirus that can form functional virus and are
infectious, and they can move around the genome. Knockout mutations
make copies of the retrovirus nonfunctional in that they can't form
infective virus and they get stuck where they are and slowly degenerate
by random mutations into random sequence.

If you pick a pseudo gene such as the GULO gene you will identify
mutations that have been pretty much exclusively neutral since they
became non functional and they would be better for the type of analysis
that you want to do.

Pretty soon there will be a lot of this sequence available for the
entire GULO pseudogene and you can work on that.

They have a start:
https://www.broadinstitute.org/scientific-community/science/projects/mammals-models/29-mammals-project

Bush babies are prosimians, and you can compare their functional
sequence to the GULO pseudogene in monkey, chimp and human. They have
the 10,000 vertebrate genome project and they intend to sequence
hundreds of primates. In a few years you will have undeniable evidence
that you are absolutely wrong, but you will likely still be in denial.

Before you do any fancy analysis you might want to think about what it
tells you that monkeys, chimps and humans share the same defective gene.
Common descent is a fact of nature. Your analysis will only
demonstrate the fact that you should already be able to figure out from
the basics of the situation.

Ron Okimoto

John Harshman

unread,
Oct 1, 2016, 1:19:56 PM10/1/16
to talk-o...@moderators.isc.org
You didn't say it. It's just implicit in your model. You seem not to
understand your own model.

>>> And one more selection pressure targeting one of the genetic loci
>>> shuts down the amplification process.
>>
>> Why?
> Because of the multiplication rule of probabilities. The evolutionary
> trajectory has become too complex and requires too many simultaneous
> mutations to improve fitness. The probabilities for this to happen
> drops to virtually nil.

Like I said: simultaneous mutations. One mutation alone doesn't improve
fitness in your model, as I described, you denied, and now reaffirm. My
only explanation for this flipflop is that you don't understand your own
assumptions well enough to see that it's a flipflop.

>>> So how do you extrapolate a
>>> scenario like the one you propose for the transformation of a reptile
>>> population into a bird population? What are the selection pressures,
>>> what are the targeted genetic loci?
>>
>> Why this obsession with reptiles?
>
> Because they are so cute and look so much like birds, except they have no feathers.

What about Caudipteryx, Sinornithosaurus, Dilong, Sinosauropteryx,
Archaeopteryx, and Microraptor? Are they birds or reptiles?



Alan Kleinman MD PhD

unread,
Oct 1, 2016, 1:34:55 PM10/1/16
to talk-o...@moderators.isc.org
Your model does not exist in reality. Give us a real, measurable and repeatable example of your model where two selection pressures are acting on the population, having some members with beneficial mutations for one selection pressure, other members having beneficial mutations for the other selection pressure and both variants able to amplify simultaneously. And then the two variants recombine to give a lineage with both beneficial mutations. Otherwise, you are only speculating. All my work is based on real, measurable and repeatable empirical examples.
>
> >>> And one more selection pressure targeting one of the genetic loci
> >>> shuts down the amplification process.
> >>
> >> Why?
> > Because of the multiplication rule of probabilities. The evolutionary
> > trajectory has become too complex and requires too many simultaneous
> > mutations to improve fitness. The probabilities for this to happen
> > drops to virtually nil.
>
> Like I said: simultaneous mutations. One mutation alone doesn't improve
> fitness in your model, as I described, you denied, and now reaffirm. My
> only explanation for this flipflop is that you don't understand your own
> assumptions well enough to see that it's a flipflop.
Give us a real, measurable and repeatable example of your scenario. Otherwise, this scenario exists only in your mind.
>
> >>> So how do you extrapolate a
> >>> scenario like the one you propose for the transformation of a reptile
> >>> population into a bird population? What are the selection pressures,
> >>> what are the targeted genetic loci?
> >>
> >> Why this obsession with reptiles?
> >
> > Because they are so cute and look so much like birds, except they have no feathers.
>
> What about Caudipteryx, Sinornithosaurus, Dilong, Sinosauropteryx,
> Archaeopteryx, and Microraptor? Are they birds or reptiles?
Did you sequence their genomes?


RonO

unread,
Oct 1, 2016, 1:54:55 PM10/1/16
to talk-o...@moderators.isc.org

Vincent Maycock

unread,
Oct 1, 2016, 2:59:56 PM10/1/16
to talk-o...@moderators.isc.org
How do you know that it's just a "few?"

And are you sure that most or all the functioning pseudogenes have a
function related to their sequence? How can you tell?

Are there multiple ways to tell?

John Harshman

unread,
Oct 1, 2016, 7:59:55 PM10/1/16
to talk-o...@moderators.isc.org
If you were thinking straight, you would realize that fitness is a
single number, a reaction of a genotype (the sum of all loci) to the
environment (the sum of all selection pressures). It doesn't matter if
there are multiple selection pressures since they are summed when
determining fitness.

So any multilocus selection would fit my scenario, as long as there are
alleles at both loci that individually affect fitness. In a your
scenarios, no locus individually affects fitness. And that's the
difference. If your math can't handle it, so much the worse for your math.

>>>>> And one more selection pressure targeting one of the genetic loci
>>>>> shuts down the amplification process.
>>>>
>>>> Why?
>>> Because of the multiplication rule of probabilities. The evolutionary
>>> trajectory has become too complex and requires too many simultaneous
>>> mutations to improve fitness. The probabilities for this to happen
>>> drops to virtually nil.
>>
>> Like I said: simultaneous mutations. One mutation alone doesn't improve
>> fitness in your model, as I described, you denied, and now reaffirm. My
>> only explanation for this flipflop is that you don't understand your own
>> assumptions well enough to see that it's a flipflop.
> Give us a real, measurable and repeatable example of your scenario. Otherwise, this scenario exists only in your mind.

Any scenario with more than one locus affecting genotype fitness will
work. Any additive trait, for example. Do you know of any additive traits?

>>>>> So how do you extrapolate a
>>>>> scenario like the one you propose for the transformation of a reptile
>>>>> population into a bird population? What are the selection pressures,
>>>>> what are the targeted genetic loci?
>>>>
>>>> Why this obsession with reptiles?
>>>
>>> Because they are so cute and look so much like birds, except they have no feathers.
>>
>> What about Caudipteryx, Sinornithosaurus, Dilong, Sinosauropteryx,
>> Archaeopteryx, and Microraptor? Are they birds or reptiles?

> Did you sequence their genomes?

Again you avoid the question. Are you saying you can't tell a bird from
a reptile without sequencing its DNA? You must have a lot of trouble in
life, if so. Don't hide. Answer.


John Harshman

unread,
Oct 1, 2016, 8:04:54 PM10/1/16
to talk-o...@moderators.isc.org
I should have said that the few *known* functional pseudogenes do seem
to have functions that depend on sequence. And you can tell because
their sequence is conserved over evolutionary time.

Now of course there might be pseudogenes with functions unrelated to
sequence. They would be hard to recognize, but I suppose knockout
experiments might do it.

gkaplan

unread,
Oct 1, 2016, 9:19:54 PM10/1/16
to talk-o...@moderators.isc.org
On 10/1/2016 5:56 AM, RonO wrote:
> Your links do not work for some reason. This is the paper that you are
> talking about.
>
> https://www.ncbi.nlm.nih.gov/pmc/articles/PMC516556/
>
> These are not pseudogenes. There are a group of 7 genes that evolved by
> gene duplication that cluster in one region of a chromosome. This just
> means that there are 7 functional genes that evolved from one progenitor
> genes and they are close to each other on a chromosome. This paper is
> talking about a gene family (the cluster of 7 genes is a gene family)
> that diverged from each other by positive selection. This just means
> that this is not neutral evolution for some of the mutations that they
> are talking about that occurred between the 7 genes. Not only that but
> multiple types of mutations are discussed and the different types of
> mutations have different rates of occurring. Any mutation rate argument
> is not going to work because you have to figure out what the selection
> intensity was and you have to separate the different types of mutations
> under discussion.

I would not call these pseudogenes in humans either. As for a mutation
rate argument, I have already coded for selection in scilab and would
incorporate this into my procedure. It is also not difficult to apply
different mutation rates at different points of the procedure.

On the other hand, if there is a gene duplication and the required
number of beneficial mutations in a particular locus is large enough,
there will be a higher probability that there will be a deleterious
mutation at that locus which would cause that gene to be non-functional.

What interested me in the siglec is that it is obvious that the
theorists are under the gun to explain a rapid evolution. They call it
the Red Queen effect with the Red Queen being from Alice and Wonderland.

It appears to me that this philosophy (not science) borders on a need
for a directed mutation to escape an adversary. But according to my
textbook mutations are agnostic to need.

Would not this be a compromise on the part of evolutionists to appeal to
the Red Queen effect?


> You would not be able to figure out mutation rates from the sequence
> itself. What you are observing are the fixed mutations between species,
> and fixation in this case could be due to positive selection, and
> different rates of mutation and so is not dependent on some general
> mutation rate like single nucleotide neutral mutations are.

Did you look at my proposed coding procedure? It will take into account
this type of variability.


> You would be just spinning your wheels and getting no where.
>
> Pick a pseudo gene that was non functional and you would do better.

That sounds like another good project.

gkaplan

unread,
Oct 1, 2016, 9:19:54 PM10/1/16
to talk-o...@moderators.isc.org
You still feeling it from our discussion of Skinner :)

gkaplan

unread,
Oct 1, 2016, 9:24:53 PM10/1/16
to talk-o...@moderators.isc.org
I see this was duplicated, but I forgot to give the link for the Red
Queen effect - > https://en.wikipedia.org/wiki/Red_Queen_hypothesis

The phenomenon's name is derived from a statement that the Red Queen
made to Alice in Lewis Carroll's Through the Looking-Glass in her
explanation of the nature of Looking-Glass Land:


Now, here, you see, it takes all the running you can do, to keep in the
same place.[11]

Van Valen coined the hypothesis “Red Queen” because under this
interpretation where species have to “run” or evolve in order to stay in
the same place or remain extant.


RonO

unread,
Oct 1, 2016, 10:29:54 PM10/1/16
to talk-o...@moderators.isc.org
Doesn't matter, wrong system. Pick another.

Ron Okimoto

RonO

unread,
Oct 1, 2016, 10:29:54 PM10/1/16
to talk-o...@moderators.isc.org
This is not true. Why would you think this? 7 genes are evolving. In
this case there is no optimum sequence because the target is always
changing. What do you not get. If these genes do not change fast
enough to identify the changing targets you won't see them because they
would be failures.

You have picked the wrong system again. All you have is a system that
is evolving fast enough for the organism to survive under the current
conditions. If this were not true the population would be extinct and
you wouldn't be observing them.

Think about it for a second or two. You really have picked the wrong
system again. It is like claiming that the adaptive immune system can't
work because it obviously does in the number of cell divisions and
mutational events that happen.

Pick another system and try again. If your simulation doesn't reflect
reality, it is your simulation that you have to fix, not reality. It is
obvious that the human lineage has survived with this system for a very
long time, so what is your beef?

As you likely know your argument is bogus for pseudogenes and you likely
should have never posted this in this thread.

Ron Okimoto

gkaplan

unread,
Oct 1, 2016, 10:44:54 PM10/1/16
to talk-o...@moderators.isc.org
So you object is that this does not fit because I posted it in a
pseudogene thread? As for the Red Queen effect, this paper does appeal
to it as does another one on the same subject.

Also, you claim that the mutations are not single point mutations and
thus have different mutation rates. However what I was looking at were
the synonymous and non-synonymous substitutions.

Do you disagree with wiki on the definition of these?

https://en.wikipedia.org/wiki/Synonymous_substitution
A synonymous substitution (often called a silent substitution though
they are not always silent) is the evolutionary substitution of one base
for another in an exon of a gene coding for a protein, such that the
produced amino acid sequence is not modified.

https://en.wikipedia.org/wiki/Nonsynonymous_substitution
A nonsynonymous substitution is a nucleotide mutation that alters the
amino acid sequence of a protein. It is contrasted with synonymous
substitution which do not alter amino acid sequences. As nonsynonymous
substitutions result in a biological change in the organism, they are
subject to natural selection.

gkaplan

unread,
Oct 1, 2016, 10:49:54 PM10/1/16
to talk-o...@moderators.isc.org
I can do both. That being said, the mutation rate for gene duplication
is said to be the same as single point mutations and the synonymous and
non-synonymous mutations are also nucleotide mutations. The deleterious
mutation rate far exceeds the putative beneficial as well.

The Red Queen cannot help you here.

I can easily use neutral drift selection for the synonymous and apply
selection to the others.


Bill Rogers

unread,
Oct 2, 2016, 6:49:54 AM10/2/16
to talk-o...@moderators.isc.org
What do you mean by neutral drift selection? When a mutation is neutral there's no selection. Drift is an alternative to selection, not a type of selection.

RonO

unread,
Oct 2, 2016, 8:59:54 AM10/2/16
to talk-o...@moderators.isc.org
If you do not know the difference, why would you think that I am not
correct in my assessment? Pseudogenes are about as far from a selection
argument as you can get. They are nonfunctional genes that were genes
at one time or in one form. If you want to talk about positive
selection you should start another thread. The pseudogene argument
against creationism is that once a gene becomes a psedogene and is
nonfunctional any new mutation is neutral. There is no reason why
chimps and humans share more mutations in common with each other than
they do with a monkey. They share the common mutations because they
share a more recent common ancestor that had those mutations before
chimps and humans ever existed.

There is no reason for chimps and humans to share common mutations in
pseudogenes when they have no effect on the defective gene.

It is a totally different argument than what you are putting up.

>
> Also, you claim that the mutations are not single point mutations and
> thus have different mutation rates. However what I was looking at were
> the synonymous and non-synonymous substitutions.

When did I claim this? Again you are likely talking about fixation
between species. non-synonymous mutations are less likely to be fixed
between species. They have the same mutation rate, but we see fewer of
them between species because they are selected against. Just look up
the data. Non-synonymous mutations are more likely to occur just by
chance. Around 2:1 non to syn. What we observe between species is a
ratio of 1:3 to 1:5 depending on the gene. Some genes are more
conserved than others. Among primates there are no non-synonymous
mutations fixed in some of the histone genes. There are only synonymous
mutations that have been fixed. This is due to the fact that a lot of
amino acid substitutions in a protein alter the function of the protein
and are selected against, so non-synonymous mutations are fixed at a
lower frequency than synonymous mutations. It is all observed and you
can look it up if you want to.

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1456207/

In pseudogenes there are no synonymous and non-synonymous mutations.
What a bonehead. The gene is already defective and any new mutation
doesn't matter. There is no functional coding sequence in a pseudogene.

>
> Do you disagree with wiki on the definition of these?
>
> https://en.wikipedia.org/wiki/Synonymous_substitution
> A synonymous substitution (often called a silent substitution though
> they are not always silent) is the evolutionary substitution of one base
> for another in an exon of a gene coding for a protein, such that the
> produced amino acid sequence is not modified.

So what? There are no non-synonymous mutations in a pseudogene because
there is no functional coding sequence. We may label the mutations as
once being non-synonymous, but that is only to demonstrate that there is
no difference created by the previous functional coding sequence of the
now defective gene. In pseudogenes non-synonymous mutations (the
mutations that would code for an amino acid if the gene were functional)
are just as likely to be fixed as synonymous. There is no selective
disadvantage like there is for a functional gene.

>
> https://en.wikipedia.org/wiki/Nonsynonymous_substitution
> A nonsynonymous substitution is a nucleotide mutation that alters the
> amino acid sequence of a protein. It is contrasted with synonymous
> substitution which do not alter amino acid sequences. As nonsynonymous
> substitutions result in a biological change in the organism, they are
> subject to natural selection.
>

You had the wrong system. This doesn't matter for pseudogenes.

Ron Okimoto

RonO

unread,
Oct 2, 2016, 9:24:52 AM10/2/16
to talk-o...@moderators.isc.org
No you can't.

> That being said, the mutation rate for gene duplication
> is said to be the same as single point mutations and the synonymous and
> non-synonymous mutations are also nucleotide mutations. The deleterious
> mutation rate far exceeds the putative beneficial as well.

What did you learn from the human genetic variation paper that I gave
you? Gene duplication is relatively rare. What does that tell you
about the rate of such mutations? How could the rates be the same if
there are 10s of millions of single nucleotide SNP and there are only on
the order of 10,000 CNV (copy number variants)? Just think for a second
and you should realize that there is a very large difference in rate.
We are not talking about a small difference by several orders of magnitude.

In the paper that you are talking about they look at repeat regions
within a gene. These repeats are in tandem and screw ups are common.
Tandem just means that one copy is put head to tail with another copy
and you can have many such copies strung together. Some of the genes
have 16 or more copies in tandem and others genes have 2 copies. Tandem
copies have high mutation rates that can be greater than 10^-5. This is
due to common replication errors where the polymerase could jump copies
or slip back to a previous copy, so you can lose or gain copies during
replication. Recombination also alters the copy number of tandem
duplications. If all the copies do not line up perfectly when you
recombine the chromosomes you can lose and gain copies. There are also
single molecule recombinations that can occur with tandem duplications
and you lose copies in this way.

There is no way that the mutation rates are the same.

>
> The Red Queen cannot help you here.

You don't know what you are talking about.

>
> I can easily use neutral drift selection for the synonymous and apply
> selection to the others.

You don't even understand that the mutation rates are different for the
mutations that they are talking about.

Just look at the 7 gene cluster. The ancestral gene was duplcated 7
times. The first duplication is relatively rare, but once you have a
tandem duplication the rate of losing or gaining more copies can occur
at a much higher rate. This is due to the fact that the increase in
copy number after the first duplication occurs by a different mechanism.
So the hard part is getting 2 copies. This is several orders of
magnitude lower than single nucleotide mutations. Once you have the
duplication it is several orders of magnitude more likely that you can
get more copies. There will be selection for or against such copy
number variation. Humans are currently duplicating starch digesting
enzymes fairly rapidly in populations that developed agriculture and
rely on cereal grains for most of their calories. Hunter gatherers have
not selected for the increase copy number of these genes.

These are just facts that you have to deal with.

Ron Okimoto

gkaplan

unread,
Oct 2, 2016, 11:39:52 AM10/2/16
to talk-o...@moderators.isc.org
I can see why this would seem to be an oxymoron, and certainly don't
attempt here to correct you, but in our textbook on page 277 I find
three kinds of "Gene genealogies and selection":

FIGURE 8.19 Gene genealogies and selection. (A) Gene genealogy for a new
allele subject to neutral drift. In this particular case, the gene shown
drifts to fixation. (B) Gene genealogy for a new allele subject to
positive selection. Here, natural selection quickly drives the new
allele to fixation. (C) Gene genealogy for an allele under balancing
selection. Here, the two alleles both persist indefinitely in a balanced
polymorphism. Adapted from Bamshad and Wooding (2003).

Bergstrom, Carl T.; Dugatkin, Lee Alan. Evolution (Second Edition) (Page
277). W. W. Norton & Company. Kindle Edition.


The reason I clarified neutral drift by modifying it with "selection" is
that without doing so it would not be clear whether or not it was
drifting to extinction (1/2N) or to fixation in 4 Ne generations.

Even that does not quite do it, because there can be both positive and
negative selection. But what I meant was to fixation.

Vincent Maycock

unread,
Oct 2, 2016, 12:09:52 PM10/2/16
to talk-o...@moderators.isc.org
John has been saying some pseudogenes have a function related to their
function, and this seems to bear him out:

https://www.ncbi.nlm.nih.gov/pubmed/14616058

Bill Rogers

unread,
Oct 2, 2016, 1:44:58 PM10/2/16
to talk-o...@moderators.isc.org
On Sunday, October 2, 2016 at 11:39:52 AM UTC-4, gkaplan wrote:
> On 10/2/2016 3:45 AM, Bill Rogers wrote:
> > On Saturday, October 1, 2016 at 10:49:54 PM UTC-4, gkaplan wrote:
> >> On 10/1/2016 7:26 PM, RonO wrote:
> >>> On 10/1/2016 8:20 PM, gkaplan wrote:
> >>>> On 10/1/2016 9:58 AM, RonO wrote:
> >>>>> post 2
> >>>>>
> >>>>> Your links do not work for some reason. This is the paper
> >>>>> that you are talking about.
> >>>>>
> >>>>> https://www.ncbi.nlm.nih.gov/pmc/articles/PMC516556/
> >>>>>
> >>>>> These are not pseudogenes. There are a group of 7 genes that
> >>>>> evolved by gene duplication that cluster in one region of a
> >>>>> chromosome. This just means that there are 7 functional
> >>>>> genes that evolved from one progenitor genes and they are
> >>>>> close to each other on a chromosome. This paper is talking
> >>>>> about a gene family (the cluster of 7 genes is a gene
> >>>>> family) that diverged from each other by positive selection.
> >>>>> This just means that this is not neutral evolution for some
> >>>>> of the mutations that they are talking about that occurred
> >>>>> between the 7 genes. Not only that but multiple types of
> >>>>> mutations are discussed and the different types of mutations
> >>>>> have different rates of occurring. Any mutation rate
> >>>>> argument is not going to work because you have to figure out
> >>>>> what the selection intensity was and you have to separate the
> >>>>> different types of mutations under discussion.
> >>>>>
> >>>>> You would not be able to figure out mutation rates from the
> >>>>> sequence itself. What you are observing are the fixed
> >>>>> mutations between species, and fixation in this case could be
> >>>>> due to positive selection, and different rates of mutation
> >>>>> and so is not dependent on some general mutation rate like
> >>>>> single nucleotide neutral mutations are.
> >>>>>
> >>>>> You would be just spinning your wheels and getting no where.
> >>>>>
> >>>>> Pick a pseudo gene that was non functional and you would do
> >>>>> better.
> >>>>>
Ahh, maybe you think it's OK to say "neutral drift selection," because there's an example of neutral drift in a Figure labeled "gene genealogies and selection." No; neutral drift is what happens in the absence of selection, and it's in the figure as a baseline.

>
>
> The reason I clarified neutral drift by modifying it with "selection" is
> that without doing so it would not be clear whether or not it was
> drifting to extinction (1/2N) or to fixation in 4 Ne generations.
>
> Even that does not quite do it, because there can be both positive and
> negative selection. But what I meant was to fixation.

It might be best to use technical terms in the standard technical way.

Mutations can go to fixation either by drift or by positive selection. They can go to extinction either by drift or by negative selection. But drift is not selection. I am quite sure your textbook never uses the phrase "neutral drift selection."

gkaplan

unread,
Oct 2, 2016, 3:54:52 PM10/2/16
to talk-o...@moderators.isc.org
That was my thought process as it seemed they were illustrating three
different kinds of selection. In the mathematical procedures I am
contemplating, the loss due to drift (1/2N) or fixation (4 X Ne) would
be used in the same way as selection to determine how many of a
particular allele would exist in a population over time.

Thank you for the correction.


>>
>>
>> The reason I clarified neutral drift by modifying it with
>> "selection" is that without doing so it would not be clear whether
>> or not it was drifting to extinction (1/2N) or to fixation in 4 Ne
>> generations.
>>
>> Even that does not quite do it, because there can be both positive
>> and negative selection. But what I meant was to fixation.
>
> It might be best to use technical terms in the standard technical
> way.
>
> Mutations can go to fixation either by drift or by positive
> selection. They can go to extinction either by drift or by negative
> selection. But drift is not selection. I am quite sure your textbook
> never uses the phrase "neutral drift selection."

True. Back to the meat of my post, I ran some numbers last night on the
frequency of gene duplication. "The Evolutionary Fate and
Consequences of Duplicate Genes by Michael Lynch and John S. Conery" state:

These results suggest a conservative estimate
of the average rate of origin of new gene
duplicates on the order of 0.01 per gene per
million years, with rates in different species
ranging from about 0.02 down to 0.002. Given
this range, 50% of all of the genes in a
genome are expected to duplicate and increase
to high frequency at least once on time
scales of 35 to 350 million years.

I used the 0.01 per gene per million years with a population of 10,000
and fixation rate for neutrals of 1/2N and estimated that a particular
gene will be duplicated in humans and be fixed every 200 million years.

That nicely falls inside the range given above by Lynch and Conery.

Obviously on this time scale, there is no chance that in 5 million years
from when TMRCA of humans putatively evolved to modern humans a gene was
duplicated, fixed and _then_ evolved into something useful.

gkaplan

unread,
Oct 2, 2016, 4:09:52 PM10/2/16
to talk-o...@moderators.isc.org
That is fine, because my analysis has nothing to do with pseudogenes.
You don't think that one can estimate how long it would take to get 7
synonymous mutations and 42 non-synonymous mutations on the same exon?
These are all substitutions to nucleotides. According to evolutionists
it is the Red Queen effect, but being chased by pathogens does not
increase the rate of mutations, does it?



RSNorman

unread,
Oct 2, 2016, 4:29:51 PM10/2/16
to talk-o...@moderators.isc.org
On Sun, 2 Oct 2016 12:53:13 -0700, gkaplan <georg....@gmail.com>
wrote:

<snip hundreds of lines because you just jumped completely past all
that>

>True. Back to the meat of my post, I ran some numbers last night on the
>frequency of gene duplication. "The Evolutionary Fate and
>Consequences of Duplicate Genes by Michael Lynch and John S. Conery" state:
>
>These results suggest a conservative estimate
>of the average rate of origin of new gene
>duplicates on the order of 0.01 per gene per
>million years, with rates in different species
>ranging from about 0.02 down to 0.002. Given
>this range, 50% of all of the genes in a
>genome are expected to duplicate and increase
>to high frequency at least once on time
>scales of 35 to 350 million years.
>
>I used the 0.01 per gene per million years with a population of 10,000
>and fixation rate for neutrals of 1/2N and estimated that a particular
>gene will be duplicated in humans and be fixed every 200 million years.
>
>That nicely falls inside the range given above by Lynch and Conery.
>
>Obviously on this time scale, there is no chance that in 5 million years
>from when TMRCA of humans putatively evolved to modern humans a gene was
>duplicated, fixed and _then_ evolved into something useful.

Do you not understand what you, yourself just wrote? Lunch and Conery
write, as you quote, "50% of all of the genes in a genome are expected
to duplicate and increase to high frequency at least once on time
scales of 35 to 350 million years." Since humans have about 20,000
genes, then we would expact about 10,000 duplications at high
frequency in 35 to 350 million years. So in 5 million years we would
expect perhaps 1/7 to 1/70 times as many, or "only" several hundred to
several thousand.

You love to toss around Lynch and Conery but what they then say
immediately after the part you quote:
"Thus, even in the absence of direct amplification of entire genomes
(polyploidization), gene duplication has the potential to generate
substantial molecular substrate for the origin of evolutionary
novelties. The rate of duplication of a gene is of the same order of
magnitude as the rate of mutation per nucleotide site."

That does not at all sound like gene duplication is terribly rare.

Once again you completely misinterpret what you read. I have told you
this time and again including on this very point. You also like to
cite isolated pieces from Zhang's paper "Evolution by gene
duplication: an update" which calculates specifically based on Lynch
and Conery's values that there should be some 1800 gene duplications
in humans since the chimp/human divergence.

Zhang then goes on to say "Although many of these duplicated genes
might have become pseudogenes, it is possible that
some acquired new functions."

You search technical research papers that you do not really understand
to find isolated fragments that seem to suit your purpose. However
when you read the full paper it is obvious that you did not properly
interpret and use the information from the piece that you quote.

And also, one more time. Whether gene duplication plays a major role,
some role, or no role at all in the human chimp divergence has nothing
whatsoever to do with the fact that all the scientific evidence from a
number of different sources shows that humans and chimps have a common
ancestor dating back at least some 6 million years and that the
entirely of the difference between our two species could easily be
accounted for by evolutionary processes at work.

The papers under consideration are
Evolution by gene duplication: an update"

http://content.csbs.utah.edu/~rogers/bio5410/Readings/Zhang-TRE-18-292.pdf
and
The Evolutionary Fate and Consequences of Duplicate Genes
http://www.indiana.edu/~lynchlab/PDF/Lynch103.pdf

Bill Rogers

unread,
Oct 2, 2016, 4:49:54 PM10/2/16
to talk-o...@moderators.isc.org
On Sunday, October 2, 2016 at 3:54:52 PM UTC-4, gkaplan wrote:
<snip older stuff>
You seem to have misread the paper. The 0.01 gene duplications per million years is for fixation of duplications, not simply for the occurrence of the duplication in an individual. I guess you skimmed over the technical stuff looking for the bottom line, and missed the point.

You should be able to see that something was wrong in your analysis because a single fixation in 200 million years is not remotely compatible with the author's conclusion that "Given this range, 50% of all of the genes in a genome are expected to duplicate and increase to high frequency at least once on time scales of 35 to 350 million years."

You have a bad habit of skimming papers you don;t understand looking for sentences that seem understandable, but then misunderstanding them because you don't understand what experiments were actually done. It's caught you again here.



>
> That nicely falls inside the range given above by Lynch and Conery.

No, it's entirely incompatible with their conclusion.

Alan Kleinman MD PhD

unread,
Oct 2, 2016, 7:34:51 PM10/2/16
to talk-o...@moderators.isc.org
John, in your scenario, must amplification of the A and B alleles have to occur in order to improve the probability for the required recombination event?
>
> >>>>> And one more selection pressure targeting one of the genetic loci
> >>>>> shuts down the amplification process.
> >>>>
> >>>> Why?
> >>> Because of the multiplication rule of probabilities. The evolutionary
> >>> trajectory has become too complex and requires too many simultaneous
> >>> mutations to improve fitness. The probabilities for this to happen
> >>> drops to virtually nil.
> >>
> >> Like I said: simultaneous mutations. One mutation alone doesn't improve
> >> fitness in your model, as I described, you denied, and now reaffirm. My
> >> only explanation for this flipflop is that you don't understand your own
> >> assumptions well enough to see that it's a flipflop.
> > Give us a real, measurable and repeatable example of your scenario. Otherwise, this scenario exists only in your mind.
>
> Any scenario with more than one locus affecting genotype fitness will
> work. Any additive trait, for example. Do you know of any additive traits?
Before you consider additive traits, try to show if you get a single trait. So far all you have done has postulated a model for obtaining and offspring with both "beneficial" alleles by recombination. I've seen an error in your formulation and I also know of empirical examples which demonstrate your model once you have made a correction to the physics and mathematics of your model. But it isn't an example of rmns.
>
> >>>>> So how do you extrapolate a
> >>>>> scenario like the one you propose for the transformation of a reptile
> >>>>> population into a bird population? What are the selection pressures,
> >>>>> what are the targeted genetic loci?
> >>>>
> >>>> Why this obsession with reptiles?
> >>>
> >>> Because they are so cute and look so much like birds, except they have no feathers.
> >>
> >> What about Caudipteryx, Sinornithosaurus, Dilong, Sinosauropteryx,
> >> Archaeopteryx, and Microraptor? Are they birds or reptiles?
>
> > Did you sequence their genomes?
>
> Again you avoid the question. Are you saying you can't tell a bird from
> a reptile without sequencing its DNA? You must have a lot of trouble in
> life, if so. Don't hide. Answer.
You haven't sequenced the genomes. I'm not interested in fossil tea leaf reading.


Alan Kleinman MD PhD

unread,
Oct 2, 2016, 7:49:51 PM10/2/16
to talk-o...@moderators.isc.org
There is no such thing as absence of selection. Or perhaps you think every replicator replicates before it dies.
>
> >
> >
> > The reason I clarified neutral drift by modifying it with "selection" is
> > that without doing so it would not be clear whether or not it was
> > drifting to extinction (1/2N) or to fixation in 4 Ne generations.
> >
> > Even that does not quite do it, because there can be both positive and
> > negative selection. But what I meant was to fixation.
>
> It might be best to use technical terms in the standard technical way.
>
> Mutations can go to fixation either by drift or by positive selection. They can go to extinction either by drift or by negative selection. But drift is not selection. I am quite sure your textbook never uses the phrase "neutral drift selection."
Wrong on that one Bill, drift occurs when there are a variety of causes of death or impairment of replication.


Alan Kleinman MD PhD

unread,
Oct 2, 2016, 7:49:53 PM10/2/16
to talk-o...@moderators.isc.org
There is always selection, either directional or random. With HIV one and two drug therapy, the selection is directional and the virus can evolve resistance to the selection pressures. With three drug therapy, the selection is random causing the population to drift or if you want, coalesce.

John Harshman

unread,
Oct 2, 2016, 8:09:50 PM10/2/16
to talk-o...@moderators.isc.org
Must? No. Would it increase the probability? Of course. Now, what
affects "amplification"? Would that be relative fitness?

>>>>>>> And one more selection pressure targeting one of the genetic loci
>>>>>>> shuts down the amplification process.
>>>>>>
>>>>>> Why?
>>>>> Because of the multiplication rule of probabilities. The evolutionary
>>>>> trajectory has become too complex and requires too many simultaneous
>>>>> mutations to improve fitness. The probabilities for this to happen
>>>>> drops to virtually nil.
>>>>
>>>> Like I said: simultaneous mutations. One mutation alone doesn't improve
>>>> fitness in your model, as I described, you denied, and now reaffirm. My
>>>> only explanation for this flipflop is that you don't understand your own
>>>> assumptions well enough to see that it's a flipflop.
>>> Give us a real, measurable and repeatable example of your scenario. Otherwise, this scenario exists only in your mind.
>>
>> Any scenario with more than one locus affecting genotype fitness will
>> work. Any additive trait, for example. Do you know of any additive traits?

> Before you consider additive traits, try to show if you get a single
> trait. So far all you have done has postulated a model for obtaining
> and offspring with both "beneficial" alleles by recombination. I've
> seen an error in your formulation and I also know of empirical
> examples which demonstrate your model once you have made a correction
> to the physics and mathematics of your model. But it isn't an example
> of rmns.

Let us put down "additive trait" as another term you apparently don't
know. What was the error?

>>>>>>> So how do you extrapolate a
>>>>>>> scenario like the one you propose for the transformation of a reptile
>>>>>>> population into a bird population? What are the selection pressures,
>>>>>>> what are the targeted genetic loci?
>>>>>>
>>>>>> Why this obsession with reptiles?
>>>>>
>>>>> Because they are so cute and look so much like birds, except they have no feathers.
>>>>
>>>> What about Caudipteryx, Sinornithosaurus, Dilong, Sinosauropteryx,
>>>> Archaeopteryx, and Microraptor? Are they birds or reptiles?
>>
>>> Did you sequence their genomes?
>>
>> Again you avoid the question. Are you saying you can't tell a bird from
>> a reptile without sequencing its DNA? You must have a lot of trouble in
>> life, if so. Don't hide. Answer.

> You haven't sequenced the genomes. I'm not interested in fossil tea leaf reading.

So you think that one can gain no information whatsoever from fossils?
Perhaps they were planted by Satan?


gkaplan

unread,
Oct 2, 2016, 8:24:51 PM10/2/16
to talk-o...@moderators.isc.org
I don't see how. My once in 200 million years is for a particular gene
to be duplicated AND fixed. That is each and ever gene will be
duplicated and fixed once in 200 million years. The article says 50%
in 35 to 350 million years. If I extrapolate my result that would be
50% in 100 million years.

I can provide you my calculation. Would you like to swap calculations?



> You should be able to see that something was wrong in your analysis
> because a single fixation in 200 million years is not remotely
> compatible with the author's conclusion that "Given this range, 50%
> of all of the genes in a genome are expected to duplicate and
> increase to high frequency at least once on time scales of 35 to 350
> million years."

I think you skipped over the word "particular" in my analysis. I am
reporting the mutation rate consistently with how the 1.1E-08 rate for
humans is handled in the papers. 1.1E-08 mutations per particular gene
per generation. That comes out to 66 per individual and 66 X population
size per generation.

In the case of gene duplication, the papers report about 15,000 genes in
the human genome. Lynch and Conery state that 50% of these 15,000 genes
will be duplicated and fixed in 35-350 million years.

I was able to break that down into separate probabilities for duplicate
rate and then apply the fixation rate to that compound number.

>
> You have a bad habit of skimming papers you don;t understand looking
> for sentences that seem understandable, but then misunderstanding
> them because you don't understand what experiments were actually
> done. It's caught you again here.
>

I believe you have misunderstood me here.

>
>>
>> That nicely falls inside the range given above by Lynch and
>> Conery.
>
> No, it's entirely incompatible with their conclusion.


It is right smack in the middle of their range. This means that if
there is a theory that a particular gene was duplicated and fixed and
then mutated to a completely new function, which is the model I was told
on this forum can provide new functionality, it would take much longer
than the entire mutation of TMRCA of humans and modern humans JUST FOR
THAT ONE NEW GENE.

>>
>> Obviously on this time scale, there is no chance that in 5 million
>> years from when TMRCA of humans putatively evolved to modern humans
>> a gene was duplicated, fixed and _then_ evolved into something
>> useful.
>

My original conclusion still stands.


Bill Rogers

unread,
Oct 2, 2016, 9:29:51 PM10/2/16
to talk-o...@moderators.isc.org
Quite right. I misunderstood your calculation.

>
>
>
> > You should be able to see that something was wrong in your analysis
> > because a single fixation in 200 million years is not remotely
> > compatible with the author's conclusion that "Given this range, 50%
> > of all of the genes in a genome are expected to duplicate and
> > increase to high frequency at least once on time scales of 35 to 350
> > million years."
>
> I think you skipped over the word "particular" in my analysis. I am
> reporting the mutation rate consistently with how the 1.1E-08 rate for
> humans is handled in the papers. 1.1E-08 mutations per particular gene
> per generation. That comes out to 66 per individual and 66 X population
> size per generation.
>
> In the case of gene duplication, the papers report about 15,000 genes in
> the human genome. Lynch and Conery state that 50% of these 15,000 genes
> will be duplicated and fixed in 35-350 million years.
>
> I was able to break that down into separate probabilities for duplicate
> rate and then apply the fixation rate to that compound number.

Yup. I misunderstood you. I think the reason I misunderstood you was based on your conclusion that the rate of fixation of duplicates was too low to be important in a time frame of 5 million years. That's why I thought you were calculating the rate for any fixed duplication at all, rather than for each possible gene.

So if you take your numbers, 50% of 15,000 (a lowish estimate of the number of human genes, but OK) is fixed in ~150 million years that gives a rate of one new fixed duplication each 10,000 years, or 500 fixed duplications in 5 million years. And note that the paper only counted functional duplicates - pseudogenes were excluded.

I don't particularly see why 500 gene duplications is obviously too few to have an effect on the evolution of chimps and humans from the LCA.

>
> >
> > You have a bad habit of skimming papers you don;t understand looking
> > for sentences that seem understandable, but then misunderstanding
> > them because you don't understand what experiments were actually
> > done. It's caught you again here.
> >
>
> I believe you have misunderstood me here.

I did misunderstand your calculation, you are correct about that, but only because your conclusion would follow better from the misunderstood version than from the correct calculation.

>
> >
> >>
> >> That nicely falls inside the range given above by Lynch and
> >> Conery.
> >
> > No, it's entirely incompatible with their conclusion.
>
>
> It is right smack in the middle of their range. This means that if
> there is a theory that a particular gene was duplicated and fixed and
> then mutated to a completely new function, which is the model I was told
> on this forum can provide new functionality, it would take much longer
> than the entire mutation of TMRCA of humans and modern humans JUST FOR
> THAT ONE NEW GENE.

Only if you specify that gene in advance. Many genes, on the order of 500, would be expected to have been duplicated and fixed. Humans are what humans are because of which ones happened to get duplicated in those 5 million years. Evolution was not aiming at humans and hoping to hit some specific gene out of 15,000 and duplicate just that one.

We're back to the enormous improbability of you having been born, considering the odds against all those specific ancestral sperm getting lucky.

RonO

unread,
Oct 2, 2016, 9:39:51 PM10/2/16
to talk-o...@moderators.isc.org
Gene duplication is not terribly rare, I would call them relatively
rare. Gene duplication is more common than we once thought before we
could sequence genomes, but it is still not a very common genetic
variant in any population. It is found at several orders of magnitude
lower frequency in the population than single nucleotide polymorphism.
The mutation rate may be the same, but there are only 15,000 genes to
duplicate while there are 3 billion base-pairs in the human genome.
This is why we only find on the order of 10,000 gene duplications among
2,500 humans while we find over 38 million single nucleotide genetic
variants segregating among these same 2,500 individuals

Gene duplications are not as common as single nucleotide variants.
Everyone may have around 60 new single nucleotide mutation in their
genomes, but they likely have no new gene duplication mutations. The
rate of duplication for any gene may be the same as for an average
base-pair, but you still have 4 or 5 orders of magnitude fewer gene
duplications per genome just because you have fewer genes than base-pairs.

Ron Okimoto

RonO

unread,
Oct 2, 2016, 9:44:51 PM10/2/16
to talk-o...@moderators.isc.org
This depends on what kind of pseudogene. There are multiple types.
There is no doubt that some function might be maintained by some
pseudogenes, they just do not have their normal function.

Beats me what this will do for him. It just means that evolution
happens. Functions change.

Ron Okimoto

rsNorman

unread,
Oct 2, 2016, 10:14:52 PM10/2/16
to talk-o...@moderators.isc.org
RonO <roki...@cox.net> Wrote in message:
>> several thousand.7
>>
>> You love to toss around Lynch and Conery but what they then say
>> immediately after the part you quote:
>> "Thus, even in the absence of direct amplification of entire genomes
>> (polyploidization), gene duplication has the potential to generate
>> substantial molecular substrate for the origin of evolutionary
>> novelties. The rate of duplication of a gene is of the same order of
>> magnitude as the rate of mutation per nucleotide site."
>>
>> That does not at all sound like gene duplication is terribly rare.
>
> Gene duplication is not terribly rare, I would call them relatively
> rare. Gene duplication is more common than we once thought before we
> could sequence genomes, but it is still not a very common genetic
> variant in any population. It is found at several orders of magnitude
> lower frequency in the population than single nucleotide polymorphism.
> The mutation rate may be the same, but there are only 15,000 genes to
> duplicate while there are 3 billion base-pairs in the human genome.
> This is why we only find on the order of 10,000 gene duplications among
> 2,500 humans while we find over 38 million single nucleotide genetic
> variants segregating among these same 2,500 individuals
>
> Gene duplications are not as common as single nucleotide variants.
> Everyone may have around 60 new single nucleotide mutation in their
> genomes, but they likely have no new gene duplication mutations. The
> rate of duplication for any gene may be the same as for an average
> base-pair, but you still have 4 or 5 orders of magnitude fewer gene
> duplications per genome just because you have fewer genes than base-pairs.
>

You are right in relative terms. We have millions of nucleotide
changes from the common ancestor with chimps yet, given the rates
above and calculations below we make have only hundreds or so
gene duplications during that same time. Gene duplications are
far less common than nucleotide variants. Still there are a lot
of them. That is 4 or 5 orders of magnitude. Whatever, there
are a "whole bunch" of gene duplictions.

It is also true that almost all gene duplications end up as dead
ends, as pseudogenes. Perhaps a tiny number of those so-called
"pseudogenes" really do have some effect on gene transcription
and regulation even if they produce no functional gene product
themselves. Perhaps an even smaller number ends up producing a
variant gene product that does something a bit different from the
original and so alters development. And perhaps yet an even
smaller number ends up producing a gene product that eventually
results in an entirely new function.

Still, there are enough gene duplications in tens and hundreds of
millions of years to seriously influence the large scale
macroevolutionary events involved in differentiating classes and
subphyla. And there is still a possibility of enough (enough
being one or two) gene duplications resulting in functional
differences to have an impact on the divergence of humans and
chimps.

No, George, I am not saying gene duplication is "the" answer. It
is merely one of a number of changes that en masse and with
cumulative changes over time result in the divergence between
humans and chimps in six million years.
--


----Android NewsGroup Reader----
http://usenet.sinaapp.com/

RonO

unread,
Oct 2, 2016, 10:19:50 PM10/2/16
to talk-o...@moderators.isc.org
You were responding to the wrong thread. Do you know what proteins that
are discussed in your paper and what they do? If you can figure out why
there are 42 non-synonoymous changes in one exon you can likely answer
your own questions about it. What does this protein family do?

If you read the paper you know that they can't differentiate the excess
of non-synonymous from chance. There is an excess, but it isn't
significant. To reach significance they put all the Ig1 domains
together into one sequence for each species. I assume they used all 7
genes, so this would be 6 substitutions per gene and 3 would be in
humans and 3 would be in chimps. Ig1 domain consists of around 380
amino acids so this would be less than 1% change in the protein sequence
in each species. What do you think is such a big deal?

QUOTE:
However, in no case did Fisher's exact test reject the null hypothesis
that the higher nonsynonymous substitution frequency compared to the
synonymous ones is due to chance. To achieve a more robust statistical
analysis, we concatenated all CD33rSiglec Ig1-coding exon in a species
and compared it with a counterpart in another species (see Methods for
details).
END QUOTE:

Your 42 substitutions in table 3 come from this data set. In the
materials and methods they claim to have concatenated all the human Ig1
sequences and did the same for chimps. Read the paper again and see for
yourself. Your 42 substitutions is really only 3 per gene in each species.

Ron Okimoto

RonO

unread,
Oct 2, 2016, 10:24:51 PM10/2/16
to talk-o...@moderators.isc.org
I agree. There are more gene duplications occurring than we once
thought, but we always knew that gene duplications were the cause of all
the numerous gene families. Having more than we thought just means that
more can result in nothing of significance, while enough result in what
we observe.

Ron Okimoto

gkaplan

unread,
Oct 2, 2016, 10:39:50 PM10/2/16
to talk-o...@moderators.isc.org
My current research deals with the "evolution of CD33-related siglecs"
which is a very specific protection against a particular pathogen. The
papers all appeal to the "Red Queen" effect. To keep ahead of this
threat there are 42 non-synonymous mutations between chimpanzee and
human in one exon.

To escape this threat there would need to be gene duplication not just
on one gene, but an amplification of an entire gene cluster. And then
after fixation there would need to be 42 beneficial mutations on just
one of the three exons listed in the paper. Since deleterious
mutations outnumber beneficial ones, what are the odds of all this
occurring in 5MY?

Slim odds.

gkaplan

unread,
Oct 2, 2016, 11:04:51 PM10/2/16
to talk-o...@moderators.isc.org
There is not just lg1, but also lg2 and lg3 each with differences
between chimps and humans. If what you say is true, how can 7
different genes all have the same exon? Because that is how they label
lg1, 2 and 3, as exons.

Also, I am not relying on just this one paper. Your explanation, which
I need to study makes it appear that there are very few changes between
chimp and humans when the chart lists 42.

But note the following:
Rapid evolution of binding specificities and expression patterns of
inhibitory CD33-related Siglecs in primates - The FASEB Journal Vol.28,
No.3 , pp:1280-1293, September, 2016

Multispecies comparisons of the CD33rSiglec gene cluster showed

-------> extensive differences between humans,
chimpanzees, baboons, and murine species, <-------

involving rapid evolution through multiple mechanisms that
range from expansions of gene subsets, gene deletions,
pseudogenization, gene conversion events, and exon
shuffling to

-----> higher rates of nonsynonymous substitutions <-------
(amino acid changes) in the V-set domain (7, 15,
33, 34).

RonO

unread,
Oct 2, 2016, 11:39:51 PM10/2/16
to talk-o...@moderators.isc.org
So what? The other exons have fewer than 1% substitutions. What do you
not get? You had the wrong idea about the 42 substitutions. It turns
out that it is no big deal.

>
> Also, I am not relying on just this one paper. Your explanation, which
> I need to study makes it appear that there are very few changes between
> chimp and humans when the chart lists 42.

42 over all the genes of the family between two species. Figure out the
numbers for yourself. It is higher than most other proteins, but that
is how they detect positive selection. If there are more amino acid
substitutions than expected, positive selection is indicated. There
just isn't the high number that you thought. In fact, it is just right
on the edge where it might not be statistically significant.

>
> But note the following:
> Rapid evolution of binding specificities and expression patterns of
> inhibitory CD33-related Siglecs in primates - The FASEB Journal Vol.28,
> No.3 , pp:1280-1293, September, 2016

I just told you that the number is higher than most other proteins
between chimps and humans, but what do you expect if there is positive
selection involved? The number of substitutions is just not as high as
you thought by an order of magnitude difference.

Give it up and try something else.

Ron Okimoto

gkaplan

unread,
Oct 3, 2016, 2:14:50 AM10/3/16
to talk-o...@moderators.isc.org
That is not true, it is probably the "most rapidly evolving loci in the
entire genome". You don't think this is statistically significant?

Recent lower-resolution comparative studies of the initial draft
sequence of the chimpanzee genome indicate that the CD33rSiglec
gene cluster may be one of the most rapidly evolving loci in the
entire genome (Tarjei Mikkelsen, personal communication). Although
sequence identity between human and chimpanzee KLKcoding
sequences (KLK4,5,7~14) is 99.2% [7,866 nucleotides (nt)],
it is 98.5% (12,267 nt) for Siglec-coding sequences (CD33 Siglec-3,
Siglec-5~10, and Siglec-12) and 98.2% (3,253 nt) for Ig1-coding
sequences.

Bill Rogers

unread,
Oct 3, 2016, 7:39:49 AM10/3/16
to talk-o...@moderators.isc.org
On Monday, October 3, 2016 at 2:14:50 AM UTC-4, gkaplan wrote:
<snip old stuff>
A couple of points. The paper is about the Siglec gene family. These genes, of which there are multiple copies in the same chromosomal region of primates and rodents are lectins, meaning they bind complex carbohydrates, like sialic acid. The binding domains have homologies to immunoglobulins - that's why the domains of the proteins are labeled Ig1, Ig2, Ig3, etc. The various Siglec genes in different species have different numbers of these immunoglobulin-like domains, each encoded by an individual exon.

In the comparison you are talking about, the authors compared the Ig1 domains from 7 different Siglec genes in humans and chimps. If you look at Figure 2, you can see how the organization of the various Siglec genes in humans and chimps (and baboons, mice and rats) lines up. You'll also notice that there are a bunch of apparently non-functional pseudogenes derived from Siglec genes in the same region.

In comparing the domains between humans and chimps the authors are not looking at mutation rates to figure out if the mutations could have accumulated in 5 million years - that's your hobby horse, not theirs. Rather they are looking at what sorts of mutations occur. They are looking for the pN/pS ratio. What that means is this. If there is no selection on a sequence, the ratio of non-synonymous mutations to the number of possible non-synonymous mutations will be equal to the ratio of the number of synonymous mutations to the number of possible synonymous mutations. In symbols pN = pS or pN/pS = 1. That happens because if the sequence is under no selective pressure it doesn't matter what mutations accumulate.

If the sequence is under purifying selection - meaning it's a functional gene and most changes in amino acid sequence would be deleterious, then you expect pN/pS < 1. And that is true for huge numbers of genes that have been compared between species, enzymes of intermediary metabolism, genes encoding structural proteins, genes encoding proteins involved in intracellular signaling and many more.

The interesting case is where pN/pS > 1. That is taken to imply that there is actually positive selection for rapid diversification. This is where the Red Queen effect and the evolutionary arms race with pathogens comes in. Lots of pathogens use interactions between complex carbohydrates and cellular surface proteins to get access to the cell. So changes in the cell surface proteins can provide a selective advantage, at least until the pathogen evolves a different surface molecule to use to bind to the cell surface. This sort of arms race can, in principle, select for rapid evolution of surface molecules, like the Ig1 domain of the Siglecs.

So the authors were interested in asking whether pN/pS > 1 for the Siglecs compared between humans and chimps. If so, that might have suggested that Siglecs were changing rapidly under selective pressure from pathogens. And the answer was inconclusive - there were some examples of pN/pS > 1, but across all the comparisons made, they were not statistically significant. The rapid evolution of the Ig1 domains in Siglecs suggests that it might be involved in host-pathogen interactions, but the strong pN/pS signal that would have been additional evidence was not found.

Now, back to your hobby horse. An Ig domain has about 250 base pairs in it. The authors looked at 7 such domains, so they are looking at about 1750 base pairs. They found 49 total mutations between chimps and humans, separated by 10 million years of evolution (5 million each for LCA to chimp and LCA to humans). That's .028 mutations per base pair over 10 million years or 2.8*10^-9 per year. Call a generation, very generously, 28 years, and you get a mutation rate on the order of 10^-8 mutations per base per generation. Nothing especially remarkable. And that's in the fastest evolving domain, Ig1, of Siglec.

RonO

unread,
Oct 3, 2016, 7:39:49 AM10/3/16
to talk-o...@moderators.isc.org
Your own reference tells you that the highest isn't that high. The 7
genes that your original reference used did not reach significance on
their own. You are only dealing with a less than two fold difference in
your example above. Think about what that means. You were an order of
magnitude off in how you were misusing the data in your original example.

Think about it. The average coding sequence is around 1,000 base-pairs.
There are smaller and larger genes, but this is just the mean. That
means that the average gene is around 330 amino acids. The mean
sequence difference between coding sequences of chimp and human is
around 0.7%. This means that there are around 7 fixed mutations between
these species half in humans and half in chimps in 1,000 base-pairs of
coding sequence. The usual range of ratios for around 70% of the genes
is 3:1 to 5:1 silent to replacement. There are still a lot of genes
that are more conserved or less conserved than this. This just means
that you have 1 or 2 amino acid substitutions in your average coding
sequence of 330 amino acids. This would be around 1 per lineage. Your
paper found an average of 3 per lineage in a 380 amino acid coding
sequence. That is why they could not tell that it was significant just
based on one 380 amino acid sequence. It could have been due to chance
because some coding sequences obviously have more than 1 substitutions
per lineage and some do not have any. The first 9 or 10 proteins that
were sequenced were identical between chimps and humans, but with that
old tech we could only sequence the shorter proteins like insulin and
hemoglobin.

Do you see why the highest is not that high?

You are just wrong. Pick something else to go on about. This issue is
pretty much dead. Your own new reference is only a 1 fold difference
(99.2 to 98.5%). Why would that make you happy to have that as the
highest yet found? We are talking about thousands of genes that have a
distribution of results just by random chance. When you throw in
selection you expect larger deviations than chance.

Ron Okimoto

gkaplan

unread,
Oct 3, 2016, 1:14:49 PM10/3/16
to talk-o...@moderators.isc.org
Thanks, that is a good summary. A couple of points.

Siglec-11 is the gene that produces the human specific protein, so
would'nt one assume that at least this one had its origin after the
divergence? If the change happened in TMRCA of both chimps and humans,
would not both show those changes too?

I found this more detailed chart of the differences between humans and
primates from the

link below and found there are 16 differences in this gene, 15
substitutions and one deletion.

This is at least 8 changes that must happen on the same gene. I might
even be able to eventually tailor the probability based on the kind of
substitution, what do you think?

Also, your mutation rate is skewed because you are assuming that each
and every mutation is beneficial. How many different mutations are
possible? There are different substitutions (e.g. G > A, G > T; T > C,
T > A; Same to Same!), insertions, deletions, etc. There are other
possibilities as well, but with what I have just listed, you would need
a mutation rate at least 5X higher than what you came up with.

---
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3408085/figure/fig5/
Counted differences between Human (Hsa) and Chimp (Ptr)

SIGLEC11 - Hsa -> Ptr differences (16)
---------------------------------------
17 A >G
47 C >T
98 G > A
134 G > A
163 A > G
164 A > G
177 A >G
232 T > C
288 C > T
312 C > T
349 C > T
351 G > A
359 C > T
365-380 Deletion
384 G > A
379 G > T









gkaplan

unread,
Oct 3, 2016, 2:09:50 PM10/3/16
to talk-o...@moderators.isc.org
Bill, I found another paper that firms up the numbers of when the above
sequences originated. Basically the time period is from 1.1 MY ago to
about 100,000 years ago (out of Africa), so all 16 changes would be in
the human population during a one million year period.

In addition to this, two tandem gene conversion happened during the same
time period.



--
Evolution of Siglec-11 and Siglec-16 Genes in Hominins - Mol. Biol.
Evol. 29(8):2073–2086. 2012

SIGLEC16P to SIGLEC11 Conversion Events
Occurred after the Emergence of the Genus Homo
but before the Emergence of Modern Homo sapiens
Analysis of the converted SIGLEC11 sequence in the current
Human Single Nucleotide Polymorphism (SNP) database
showed no evidence for gene conversion polymorphisms
(data not shown). Taken together with the finding that
Siglec-11 is expressed in all human brains studied, the gene
conversion of SIGLEC11 is apparently universal to modern
humans (i.e., was fixed prior to the emergence of modern
humans in Africa ;100–200,000 years ago). We took advantage
of the evolutionary neutrality of the pseudogene
allele hSIGLEC16P to date the gene conversion events. After
the SIGLEC16P / SIGLEC11 gene conversions, mutations
in SIGLEC16P should have accumulated at a neutral rate.
The timing (T) of gene conversion can therefore be roughly
calculated by T 5 d/k, where d is the branch length of hSIGLEC16P
and k is the neutral mutation rate of genomic
region containing SIGLEC11 locus. Under the assumption
of 6 myr of human–chimpanzee divergence, k was calculated
as (1.4 ± 0.1) 10 9/site/year using the standard
human and chimpanzee genetic distance (0.0166). The
d value is estimated as 0.0016 in A/A#1 þ A/A#2 (based
on the number of changes that occurred only in human
SIGLEC16P). Thus, T is estimated as 1.1 myr.

Bill Rogers

unread,
Oct 3, 2016, 2:09:50 PM10/3/16
to talk-o...@moderators.isc.org
Sorry, don't follow you here. According to the figure you just linked to, Siglec-11 is present in humans, chimps, and gorillas.

>
> I found this more detailed chart of the differences between humans and
> primates from the
>
> link below and found there are 16 differences in this gene, 15
> substitutions and one deletion.
>
> This is at least 8 changes that must happen on the same gene. I might
> even be able to eventually tailor the probability based on the kind of
> substitution, what do you think?

I think you are wasting your time. You are calculating as though the human sequence was a goal chosen in advance.

>
> Also, your mutation rate is skewed because you are assuming that each
> and every mutation is beneficial. How many different mutations are
> possible? There are different substitutions (e.g. G > A, G > T; T > C,
> T > A; Same to Same!), insertions, deletions, etc. There are other
> possibilities as well, but with what I have just listed, you would need
> a mutation rate at least 5X higher than what you came up with.

The rate I gave is the observed rate of fixed mutations, whether fixed by drift or selection. You are still calculating as though the human sequence as a target chosen in advance.

If you are going to do that, just imagine all the genome-wide differences between humans and chimps. The probability of the human sequence arising with 5 billion years, never mind 5 million, is miniscule.

Bill Rogers

unread,
Oct 3, 2016, 4:04:49 PM10/3/16
to talk-o...@moderators.isc.org
It's the tandem gene duplications that are likely to be evolutionarily important.
A very interesting paper. Shows how pseudogenes can provide raw material for evolution via gene conversion.

I think you have the wrong idea about the timing of the individual mutations though. This paper is tricky, and you need to understand gene conversion well to understand the paper.

[Gene conversion is covered only briefly in your text, but you can read about it here..... https://en.wikipedia.org/wiki/Gene_conversion the specific sort of gene conversion at play in your paper is known as ectopic gene conversion]

The sequence in hsaSIGLEC11 comes from the sequence hsaSIGLEC16P, at least for the parts of the gene that underwent the gene conversion event(s) 1-1.2 million years ago. You can see a map of where the gene conversion occurred in figure 1A of the paper.

That sequence from hsaSIGLEC16b that replaced the original hsaSIGLEC11 sequence in the gene conversion was already different from the chimp sequence from before the time of the gene conversion events 1-1.2 million years ago. The sixteen differences did not accumulate in just the 1 million years after the gene conversion.

gkaplan

unread,
Oct 3, 2016, 4:24:50 PM10/3/16
to talk-o...@moderators.isc.org
Why is siglec-11 found in other primates if it was developed in humans
after divergence?

Note the first line of the Abstract:

We previously reported a human-specific gene conversion of SIGLEC11 by
an adjacent paralogous pseudogene (SIGLEC16P), generating a uniquely
human form of the Siglec-11 protein, which is expressed in the human
brain. Here, we show that Siglec-11 is expressed exclusively in
microglia in all human brains studied—

Also look at figure 10 where SIGLEC11 was duplicated ~ 20mya before
SIGLECT16 emerged. Then 3-4 mya there was parallel convertion of 1)
SIGLEC16 by SIGLEC11 in humans and convertion of Hominin SIGLEC11 by
SIGLEC16P.


The explanation given to account for the facts of these species that
have the siglec gene cluster is convoluted and appeals to special
circumstances. There is a term used to describe this when the same
appeal is made in a theological explanation. It is called hapax
legomena, or a thing only said once. When there is a particular
Greek word used only one time in the Greek New Testament, and someone
builds a key argument on its usage, that is special pleading.

What are the clues that this is happening in this scientific proof?
Just look for extraordinary words and special appeals, such as the only
and first, unusual and remarkable!

Pseudogene donor conversion, FIRST time it ever happened in humans.
-------------------------------------------------------------------------------------------------
The only other vertebrate case of this is "the chicken IgY locus" and
"To our knowledge, the only other vertebrate case of a pseudogene donor
conversion that results in an intact and novel gene is the
SIGLEC16P/SIGLEC11 conversion event we studied here. Moreover, this
appears to be the
first instance of human-specific productive gene conversion event
involving a pseudogene donor." - Evolution of Siglec-11 and Siglec-16
Genes in Hominins, page 2083.

Not just one, but two tandem gene conversions, its REMARKABLE!
-------------------------------------------------------------------------------------------------
Remarkably, despite the fact that hSIGLEC16P accumulated mutations at a
neutral
rate, two tandem gene conversions delivered relatively large sequence
changes into human SIGLEC11. - ibid page 2083

SIGLEC11 survived unusual events!
-----------------------------------------------------
The SIGLEC11 gene in humans clearly survived an unusual series of
complex gene conversion events to maintain an active reading frame.
-ibid 2084
It is loading more messages.
0 new messages