Genetic basis for human evolution

[mark kennedy]

First of all, you have to understand that this is just a comparision of sequences. I was going to wait for Kerrmetric but he is going to be a while and I'm running out of time. I'm going to give you two the basic facts as I understand them. The one, perhaps crucial fact set I have ignored is the murids that they used as an outgroup. The murids were especially interesting because the mice and rats supposedly split about the same time chimpanzees and humans did. I have no interest in that, I am only interested in the direct comparisons of the chimpanzee and human genomes.

What they discovered was 35 million single nucleotide polymorphism SNP, totaling 1.23% of the differences. 14-22% were due to polymorphisms which means at least 1.06 are fixed. There were found 5 million insertions/deletions (indels), totaling 3% of the differences. There were found ~ 70,000 insertions > 80 base pairs bps in length. There were 163 regions > 15 kilobases Kb totaling 8.3 Mb (Mega bases) in the human sequence. There were exons of known genes in 34 regions.

Observable insertions into 2 classes:

1)Completely covered insertions within continuous sequence in both species.

2)Incompletly covered insertions, occuring within sequence containing one or more gaps in the chimpanzee, be tevealed by a clear discrepancy between the species in sequence length.

Indels 1 base to 15 Kb in length:

Differences in the chimpanzee-specific sequence 35 Mb.
Differences in the human-specific sequence 32 Mb.

Total divergance between human and chimpanzeee geneomes 40-45 Mb in the respective genomes which come to 90 Mb or 3%.

There are 9 larger scale rearrangements, these pericentric inversions have 18 breakpoints identified. They range from 2-4 Mb in length and total just under 20 Mb.

Gene evolution:

In all, 13,454 pairs of human chimpanzee genes were compared. Since it is assumed that the respective genomes evolved from a common ancestor, they assigned a number for each gene estiamted.

The Ka/Ks ratio is the classic measure of the overall evolutionary constraint on the gene: Ka represent the coding bases substitutions that result in amino acid change as a fraction of all such possible sites (the non-synonymous substution rate). Ks the synomomous substitution rate does not change amino acid sequences.

The Ka/Ks ratio < 1 indicates a substantial proportion of amino acid changes have been eliminated by purifying selection. Assuming Ks is neutral Ka/Ks > 1 implies, but is not necessary for adaption or positive selection.

Ok, that is not all of it but it should give us something substantive to talk about. There are your raw numbers shernern and glaudys if you want to correct or expand on anything this would be a good time.

Some of the thing that bother me about this. we have been told for decades the DNA is 98% identical in the two lineages. This was found to be false:

"For almost 30 years [from 1972], researchers have asserted that the DNA of humans and chimps is at least 98.5% identical. Now research reported here last week at the American Society for Human Genetics meeting [In Baltimore, October 2002] suggests that the two primate genomes might not be quite as similar after all. A closer look has uncovered nips and tucks in homologous sections of DNA that were'nt noticed in previous studies."

(Jumbled DNA Separates Chimps and Humans, Science Oct. 2002)

Mind you, the Chimpanzee genome is 200 Mb longer then the Human genome to begin with. Perhaps that is what they are talking about when the say Orthologous or homologous, I'm not really sure. So in the human lineage since the split we have accumulated 45Mb of indels lets say lets say 18 Mb of SNPs and 10 Mb of chromosomal rearrangements. We round it off to 70 Mb and stretch it out to the 7 million years mentioned in the paper and you have a mean of 1 nucleotide per year, 20-25 per generation for 7 million years, then it suddenly stops or did this in spurts.

In human populations the DNA and I do mean the entire genome, diverges by 1/10 of one percent. In dogs with 30,000 distinct breeds/species they only diverge by 1/10 of 1%.

I will await you response and then we can talk some more.

Grace and peace,
Mark

 

[shernren]

I'll have to tag along behind gluadys on this one, I'm no good with genetics. But I don't get your fixation on:

we have been told for decades the DNA is 98% identical in the two lineages. This was found to be false:

"For almost 30 years [from 1972], researchers have asserted that the DNA of humans and chimps is at least 98.5% identical. Now research reported here last week at the American Society for Human Genetics meeting [In Baltimore, October 2002] suggests that the two primate genomes might not be quite as similar after all. A closer look has uncovered nips and tucks in homologous sections of DNA that were'nt noticed in previous studies."

(Jumbled DNA Separates Chimps and Humans, Science Oct. 2002)

This smacks of the same idea of "fundamental distrust of science" that lies behind creationists who quote Piltdown Man and Archeoraptor as evidence that evolution is false and I'm quite sure you're above that sort of nonsense.

 

[mark kennedy]

I'll have to tag along behind gluadys on this one, I'm no good with genetics. But I don't get your fixation on:



This smacks of the same idea of "fundamental distrust of science" that lies behind creationists who quote Piltdown Man and Archeoraptor as evidence that evolution is false and I'm quite sure you're above that sort of nonsense.

We are told that we are virtually identical to chimpanzees in our DNA but we are not. We are 94% simular and only about 29% identical. A couple of nucleotides here and there, a change of environment, a couple of choice bottlenecks and bingo, Homo sapien sapein. The truth is coming out, it's not that simple. You know what really makes me smile, evolutionists act as if its no big supprise how wide the divergance rate is. Of course they were surprised and I have watched them squirm over the findings of this research paper like a bug on a hot plate.

Piltdown was a fraud, Archeorator was a bird and us being 98% ape has been conclusively proven to be false. Just as the creation science model of independant creation predicted.

This smacks of the same idea of "fundamental distrust of science"

Distrust science!? Heck no! I was counting on it and it came through with flying colors!

Grace and peace,
Mark

 

[gluadys]

Piltdown was a fraud, Archeorator was a bird

Actually Archeoraptor was a fraud too. It was two partial feathered dinosaur fossils put together as if they were one whole fossil.

 

[gluadys]

The one, perhaps crucial fact set I have ignored is the murids that they used as an outgroup. The murids were especially interesting because the mice and rats supposedly split about the same time chimpanzees and humans did. I have no interest in that, I am only interested in the direct comparisons of the chimpanzee and human genomes.

You should consider them. They indicate that evolution was happening in the chimp/human line at a much more rapid rate than in the murid lineages over the same time frame. They also provide part of the control evidence that the changes were adaptive.

The Ka/Ks ratio is the classic measure of the overall evolutionary constraint on the gene: Ka represent the coding bases substitutions that result in amino acid change as a fraction of all such possible sites (the non-synonymous substution rate). Ks the synomomous substitution rate does not change amino acid sequences.

The Ka/Ks ratio < 1 indicates a substantial proportion of amino acid changes have been eliminated by purifying selection. Assuming Ks is neutral Ka/Ks > 1 implies, but is not necessary for adaption or positive selection.

I hope you are taking note that a Ka/Ks ratio >1 "is not necessary" for adaptation or positive selection. The Ka/Ks ratio may be less than 1 and still indicate adaptive selection.


Note to shernren. This Ka/Ks ratio is based on the fact that amino acids are represented in DNA by sequences of three base nucleotides. The four base nucleotides can be arranged in 64 such sequences. But they code for 20 amino acids (+ "stop"). So there is plenty of redundancy. Most amino acids have two DNA sequences coding for them, and some have up to six. So you can easily get a change in the DNA sequence that leads to the same amino acid and therefore no difference in the protein product. That is a synonymous mutation, and since is has no effect on the protein, it is assumed to be a neutral mutation, not affected by natural selection. If it becomes fixed in the genome, it is by genetic drift not selection.

Since non-synonymous mutations can also be neutral, some of them will be fixed by chance rather than selection as well. It is not possible to sort out precisely which non-synonymous changes in a gene are adaptive rather than neutral, but the Ka/Ks ratio gives a rough idea of whether some of the fixation was due to adaptive pressure.

Ok, that is not all of it but it should give us something substantive to talk about. There are your raw numbers shernern and glaudys if you want to correct or expand on anything this would be a good time.

Why would I quarrel with the numbers. They are copied from the paper. I know you did not make them up.

But I don't understand why your conclusions are so different from the conclusions of the researchers. What do you think they missed?

Some of the thing that bother me about this. we have been told for decades the DNA is 98% identical in the two lineages. This was found to be false:

How false? I can't see being upset if it changes by a few decimal points or even a full percentage point. No one has ever claimed that science is perfect. New tools, new studies, often turn up new information. So why is this a big deal?

Also the 98% figure was never used of the whole genome, but only of coding sequences.

We round it off to 70 Mb and stretch it out to the 7 million years mentioned in the paper and you have a mean of 1 nucleotide per year, 20-25 per generation for 7 million years, then it suddenly stops or did this in spurts.

Well, a mean is just that. It doesn't imply a steady rate. Why would you not expect spurts and slow-downs?

In human populations the DNA and I do mean the entire genome, diverges by 1/10 of one percent. In dogs with 30,000 distinct breeds/species they only diverge by 1/10 of 1%.

And your point?

What do you conclude from these figures? And why?


Unlike shernren who needs numbers to understand things, numbers make my eyes glaze over and my mind go numb. You really have to take my hand and explain what they mean to you. Otherwise, my reaction tends to be "so what?"

PS I hope you can take the time to respond to my longer posts.

 

[shernren]

Note to shernren. This Ka/Ks ratio is based on the fact that amino acids are represented in DNA by sequences of three base nucleotides. The four base nucleotides can be arranged in 64 such sequences. But they code for 20 amino acids (+ "stop"). So there is plenty of redundancy. Most amino acids have two DNA sequences coding for them, and some have up to six. So you can easily get a change in the DNA sequence that leads to the same amino acid and therefore no difference in the protein product. That is a synonymous mutation, and since is has no effect on the protein, it is assumed to be a neutral mutation, not affected by natural selection. If it becomes fixed in the genome, it is by genetic drift not selection.

Since non-synonymous mutations can also be neutral, some of them will be fixed by chance rather than selection as well. It is not possible to sort out precisely which non-synonymous changes in a gene are adaptive rather than neutral, but the Ka/Ks ratio gives a rough idea of whether some of the fixation was due to adaptive pressure.

Okay, I got that. Googled the Ka/Ks thing.

Unlike shernren who needs numbers to understand things, numbers make my eyes glaze over and my mind go numb. You really have to take my hand and explain what they mean to you. Otherwise, my reaction tends to be "so what?"

The reason I'm asking for numbers is because numbers can only be interpreted in the context of other numbers. Like Scripture, numbers when taken out of context can be massaged to prove anything you want to prove (and it's a lot easier with numbers, too). For example, I could tell you that "50 Malaysian soldiers died on service in Iraq" and you'd go "no big deal" ... unless I told you that "Malaysia sent 100 soldiers". Context determines the significance of the numbers, and for me at least context means other numbers.

I get the dog divergence example. Question: how long did it take for that 0.1% to be achieved? After all, we are arguing about rates of fixation, right?

Just as the creation science model of independant creation predicted.

Did creation scientists ever say before the fact that they expected the divergence to be more than 2%? Or are they claiming it supports them after it has been discovered? The latter does not equal "predicted".

 

[mark kennedy]

You should consider them. They indicate that evolution was happening in the chimp/human line at a much more rapid rate than in the murid lineages over the same time frame. They also provide part of the control evidence that the changes were adaptive.

It is not my point of interest, I am focused on human evolution. I understand how they were using it for an outgroup and didn't find it important to the point I was trying to make.

I hope you are taking note that a Ka/Ks ratio >1 "is not necessary" for adaptation or positive selection. The Ka/Ks ratio may be less than 1 and still indicate adaptive selection.

Of course I realize that, that is how I think most adaptations happen anyway.

Why would I quarrel with the numbers. They are copied from the paper. I know you did not make them up.

I thought I might have made some fundamental errors, glad to see I got it right for once.

But I don't understand why your conclusions are so different from the conclusions of the researchers. What do you think they missed?

What they missed was the fact that it would have taken a substitution rate that is impossible. I talked with one of them, he tried to persuade me not to bring that up again. I only promised not to use it on him, he was such an interesting person I didn't want him to stop posting to the thread.

How false? I can't see being upset if it changes by a few decimal points or even a full percentage point. No one has ever claimed that science is perfect. New tools, new studies, often turn up new information. So why is this a big deal?

You have to love it, science does correct itself. Sometimes it take generations but it is going to take alot of back peddling for them to get out of this one. There is no way we became apes at the constant rate of substitutions observed in nature.

Also the 98% figure was never used of the whole genome, but only of coding sequences.

That is just plain wrong, I won't even dignify that remark with a refutation. It was going on for half a century, there is no way it can be intelligently argued against.

Well, a mean is just that. It doesn't imply a steady rate. Why would you not expect spurts and slow-downs?

Ok, ok, you mean that every 100 or so generations a hundred lines of code were fixed? Got it, that makes it much more managable.

And your point?

Been made repeatedly.

What do you conclude from these figures? And why?

My conclusion is that the divergance rate is too high to be accounted for by any known genetic mechanism. The rate is too high to be accounted for by anything short of Artemise herself intervening.

Unlike shernren who needs numbers to understand things, numbers make my eyes glaze over and my mind go numb. You really have to take my hand and explain what they mean to you. Otherwise, my reaction tends to be "so what?"

PS I hope you can take the time to respond to my longer posts.

Rest assured I respond to every post you present to me as best I can. I broke the numbers down as best I could and the raw data is here for anyone to explain as they see fit. You can say 'so what' but ultimatly that is an argument from incredulity. It has been proven what the divergance rate is, what has not been proven is how it got there.

We are not related to chimpanzees, they are in the process of proving that impossible. You say so what? I say, so the single common ancestor model just broke at the first link. We are not apes and after that the rest is silence.

Grace and peace,
Mark

 

[gluadys]

What they missed was the fact that it would have taken a substitution rate that is impossible.

Well, that's what I am looking for---some data to show that it's impossible. It would help to have some idea of what IS possible before claiming that something is impossible.

That is just plain wrong, I won't even dignify that remark with a refutation. It was going on for half a century, there is no way it can be intelligently argued against.

Sure it has been. But I have never seen the figure used, at least in scientific sources, without the qualification that it was referring to coding DNA. Journalists may have omitted that qualification in news articles.

Ok, ok, you mean that every 100 or so generations a hundred lines of code were fixed? Got it, that makes it much more managable.

Possibly, why not? There is no reason why the amount of code fixed be the same in every generation, as long as it gives the mean that has been established.

Been made repeatedly.

Well, repeat it once more please, because I still don't get it.

My conclusion is that the divergance rate is too high to be accounted for by any known genetic mechanism. The rate is too high to be accounted for by anything short of Artemise herself intervening.

I know that. What I have not seen is the data that supports that conclusion.

It has been proven what the divergance rate is, what has not been proven is how it got there.

And it has not been proven that it is impossible either.

 

[gluadys]

The reason I'm asking for numbers is because numbers can only be interpreted in the context of other numbers. Like Scripture, numbers when taken out of context can be massaged to prove anything you want to prove (and it's a lot easier with numbers, too). For example, I could tell you that "50 Malaysian soldiers died on service in Iraq" and you'd go "no big deal" ... unless I told you that "Malaysia sent 100 soldiers". Context determines the significance of the numbers, and for me at least context means other numbers.

Yes, that is really what I meant by "taking me by the hand and explaining". In this context any number up to 100 is possible. It would only be a report that 101 + Malaysian soldiers died in Iraq that would not be possible.

I get the dog divergence example. Question: how long did it take for that 0.1% to be achieved? After all, we are arguing about rates of fixation, right?

And even if one did know the rate in dogs, it would not necessarily be relevant to humans. You would have to show that the environmental pressures to adapt were of the same intensity. That's the point of the murid control in this situation. Rats and mice diverged about the same length of time ago as chimps and humans, but the degree of divergence is less, at least in the changes in the brain, presumably because there was less environmental pressure on them than on humans.

In dogs, artificial breeding has produced very intense selective pressure for certain traits. And that has been over the relatively brief time since humans domesticated them.