Evolution as natural history is psuedo-science

[busterdog]

It would be nice if you guys could find a few points of agreement and then move on from there. It would put a sharder focus on where the division actually lies. There comes a point when you just bracket all statements like, "We already disproved that" and proceed to look for so datat to get ahold of.

I also don't know if the continuing injection of new terms helps your arguments, but it makes this difficult for the lay person to follow.

Kindly indulge us logical positivists. (See, new term!)

 

[mark kennedy]

This is wrong -- Mark has misunderstood something he read. Within the modern human population, one particular class of indels (transposons) accounts for 10-20% as many variable bases as single base substitutions (SNPs). Transposons are, however, only 0.5% of all indels. Combined, all indels account for many more variable bases than do SNPs.

I just love how I'm allways misunderstanding this stuff when it's not really all that hard:

"Transposons and transposon-like repetitive elements collectively occupy 44% of the human genome sequence. "

In terms of base pairs SNPs amount to much more of the variation.

"Therefore, with 10 million bases of variation, SNPs account for the majority of common human genetic variation, followed by indels and then transposon insertion polymorphisms."

That is not ambiguise and it does not take a PHD to understand it. SNPs are the majority of variation.

"On the other hand, if we assume that the average transposon polymorphism in humans is ∼500–1000 bp in length, then the total amount of variation caused by common transposon insertions is 1–2 million base pairs (equivalent to 10–20% of the base pair variation caused by SNPs). "

10-20% of the base pair variation the 1-2 million base pair insertions.

"Thus, in terms of the number of base pairs, common transposon insertions cause significant levels of human genetic variation."

Here's the link if anyone is interested:

Natural Genetic Variation Caused by Transposable Elements in Humans



"In "An initial map of insertion and deletion (INDEL) variation in the human genome" (Genome Research 2006 16:1182-1190) the authors found 3.3 million SNPs and 534,000 indels in a survey of variation in the genome. That is, indels are 16% as common as SNPs in modern humans. Recall from the chimpanzee genome that when comparing humans and chimpanzees, indels are 14% as common as single-base substitutions. Quite consistent, I'd say, and a good indication that the human/chimp indel differences are the result of mutation.

There were 35 million base pairs and 5 million indels if that's what you are talking about. The indels come to 90 Mb which makes them over 3x bigger in terms of base pairs then single nucleotide substitutions. In the paper under discussion they are clear that they are measuring the divergance in terms of base pairs. I understand that much and dispite the constant insistant that I don't understand I think I understand this just fine.

As for the total number of bases involved in indels, we still do not have a very good estimate.

The paper suggests otherwise:

"Our method was highly efficient and led to the identification of 605 nonredundant transposon insertion polymorphisms in 36 diverse humans. We estimate that this represents 25–35% of ~2075 common transposon polymorphisms in human populations. "

The chimpanzee genome paper found that indels accounted for 2.3 - 2.6 times as many bases (different between the two species) as single-base substitutions.

35/90 Mb so that sounds about right.

The indel map cited above found in modern human variation that indels accounted for 1.3 times as many bases as single-base substitutions. That study was limited to indels shorter than 10,000 base pairs, however, and is therefore a lower limit on the total number. Indels are known to be as long as hundreds of thousands of base pairs; even small numbers of these large indels contribute enormously to the total number of bases affected by indels. When a complete survey can be done, finding that indels contribute at least 2.5 times the variation (in base pairs) as SNPs will be consistent with everything we've seen so far. (In fact, I suspect we will find that indels contribute a larger fraction to modern variation than they do to between-species differences, since they are somewhat more likely to be deleterious and thus not to stick around.)

In any case, it is clear that the number of bases that vary between humans as a result of indels is larger than the number that vary because of SNPs. In short, Mark's entire line of argument here is wrong.

I don't know that I am totally convinced of this any more then I am that mutation rates are measured in terms of base pairs. I'm not convinced that there is a point being made here, looking at the paper itself would be the best advice I could give the casual reader.

 

[sfs]

I just love how I'm allways misunderstanding this stuff when it's not really all that hard:

I find it quite sad, myself.

"Transposons and transposon-like repetitive elements collectively occupy 44% of the human genome sequence. "

Yes. And . . . what?

In terms of base pairs SNPs amount to much more of the variation.

"Therefore, with 10 million bases of variation, SNPs account for the majority of common human genetic variation, followed by indels and then transposon insertion polymorphisms."

That is not ambiguise and it does not take a PHD to understand it. SNPs are the majority of variation.

It's in fact quite ambiguous, and badly written. As the group's own numbers show, SNPs account for the majority of genetic variation as measured by events, not as measured by base pairs.

"On the other hand, if we assume that the average transposon polymorphism in humans is ?500–1000 bp in length, then the total amount of variation caused by common transposon insertions is 1–2 million base pairs (equivalent to 10–20% of the base pair variation caused by SNPs). "

10-20% of the base pair variation the 1-2 million base pair insertions.

10-20% of the base pair variation is caused by transposons. Transposons are only one half of one percent of all indels. How many base pairs do the other 99.5% of indels represent, Mark? Just answer that question.

To answer it, you should look at the paper I quoted. It's from the same group that produced the transposon paper you're quoting from. First they wrote a paper on the transposons, which contribute 10-20% as much as SNPs to base pair variation, and then they wrote a second paper about the other 99.5% of indels. The second paper makes it clear that indels taken as a whole contribute more base pairs of variation than do SNPs. Those are the facts. Interpret them however you wish, but stop getting the facts wrong.

The paper suggests otherwise:

"Our method was highly efficient and led to the identification of 605 nonredundant transposon insertion polymorphisms in 36 diverse humans. We estimate that this represents 25–35% of ~2075 common transposon polymorphisms in human populations. "

Since transposons continue to be only a tiny fraction of all indels, your comment does not respond to my point. Transposons have been well measured. Long indels (which are much longer than transposons) have not been well measured. Therefore the total amount of base pairs introduced by indels has not been well measured. Even with the measurements that we do have, however, we can set a lower limit on it that is more than ten times higher than you claim is the case.

I don't know that I am totally convinced of this any more then I am that mutation rates are measured in terms of base pairs. I'm not convinced that there is a point being made here, looking at the paper itself would be the best advice I could give the casual reader.

I don't give a fig whether you are convinced by this or by anything else. Believe that your head is a turnip, for all I care. I only respond to you so that readers won't be confused by the falsehoods you post. My advice as a geneticist is for the casual reader to ignore anything and everything that Mark says about genetics.

 

[shernren]

I was going to wait until Mark does his "expositive review of six papers" argument and go through them one by one but since sfs jumped the gun I might as well jump in. If you take a careful look at the paper it's obvious that in terms of base pairs, indels outnumber SNPs in their contribution to human natural genetic variability.


First, the section actually cited by mark:

Together with previous studies, our analysis indicates that SNPs, indels, and transposon insertion polymorphisms represent significant sources of genetic variation in humans. Human populations are estimated to harbor ∼10 million common SNPs (Judson et al. 2002), ∼2 million common indels (our unpublished data), and ∼2000 common transposon insertion polymorphisms (this study). Therefore, with 10 million bases of variation, SNPs account for the majority of common human genetic variation, followed by indels and then transposon insertion polymorphisms. On the other hand, if we assume that the average transposon polymorphism in humans is ∼500–1000 bp in length, then the total amount of variation caused by common transposon insertions is 1–2 million base pairs (equivalent to 10–20% of the base pair variation caused by SNPs). Thus, in terms of the number of base pairs, common transposon insertions cause significant levels of human genetic variation. Moreover, humans also are likely to harbor >10 million rare private transposon insertions (cases in which only one or a few individuals have the insertion). Therefore, transposon insertion polymorphisms cause significant levels of human variation.​

(emphasis added)

Pay attention to the bolded phrase: it signals that the argument has changed. "In terms of the number of base pairs", common transposon insertions cause significant levels of human genetic variation - even though SNPs outnumber transposon insertions roughly 5,000 to 1. Clearly this means that the preceding analysis ("SNPs account for the majority of common human genetic variation" etc.) was not in terms of the number of base pairs but in terms of the number of events.


This is semantics, mark, so criticize it as semantics if you will but there's more. This requires a close look at the methodology section (in which I get lost too). But first, a look at the abstract:

We began by identifying 606,093 insertion and deletion (indel) polymorphisms in the genomes of diverse humans. We then screened these polymorphisms to detect indels that were caused by de novo transposon insertions.​

"606,093 indel polymorphisms", aight? Just how long were those indels?

After the traces were successfully mapped to unique genomic locations, they were unmasked and aligned to their assigned genomic locations using the Bl2Seq program (NCBI). The Bl2Seq program allowed for as much as a 16-base gap in the alignments and led to identification of indels as large as 16 bases in length.​

A new algorithm also was developed to identify indels that were >16 bp in length. Our strategy was designed to split trace data into two blocks upon encountering a region in the pairwise alignment that no longer matched the query. The first block of sequence that matched was maintained in the correct position, and the nonmatching sequence was moved over as a block, 1 base at a time, until a match was obtained. The Perl program that was developed to accomplish this task moved the nonmatching block until it detected either a perfect alignment or a distance of 10,000 bases (the maximum distance allowed by the program). The 5 bases on each side of an indel candidate were required to have Phred scores of ≥20 to ensure that high-quality bases were being used to locate the indel junctions. Indel candidates were deposited into dbSNP under accession nos. ss8029278–ss8176133, ss8475737–ss8484870, ss14926095–ss15354938, and ss15357378–ss15378640.​

Now, I'm a bit fuzzy on the Bl2Seq program (I'm an armchair geneticist ) but clearly their settings enabled them to identify indels with length 16 base pairs and above. The candidate indels were further winnowed with the Perl program mentioned in the second paragraph to find genuine and extremely convincing indels - this means that conceivably there were more indels longer than 16bp than detected. The indel sequences were catalogued, and if you look at the accession numbers you will note that 606,093 entries were added to the database - matching the number of indels noted in the abstract.

So, to summarize what we have so far:

1. The team searched out all indels with length more than (not sure about "and equal to") 16bp.
2. They found 606,093 such indels.

What does this tell us? 606,093 indels x 16 bp/indel gives us a minimum of 9.7 million base pairs. In other words, indels contribute at least 9.7 million base pairs of variation to the human genome. The ~10 million SNPs only contribute ~10 million base pairs, so this leaves indels and SNPs almost neck-to-neck.


Is this over? No way. Jumping down to the conclusion:

Together with previous studies, our analysis indicates that SNPs, indels, and transposon insertion polymorphisms represent significant sources of genetic variation in humans. Human populations are estimated to harbor ∼10 million common SNPs (Judson et al. 2002), ∼2 million common indels (our unpublished data), and ∼2000 common transposon insertion polymorphisms (this study). Therefore, with 10 million bases of variation, SNPs account for the majority of common human genetic variation, followed by indels and then transposon insertion polymorphisms.​

Note ~2 million common indels, out of which this study publishes the identification of roughly 600k. This means that there are still 1.4 million common indels with length < 16bp. And that's a lot. Assuming that the mean length of these indels is 8bp, the 1.4 million indels add up to 11.2 million base pairs - meaning that indels, giving in total 20.9 million base pairs, outnumber SNPs' 10 million base pairs in terms of base pairs 2 to 1.

And 20.9 million base pairs is something like an absolute minimum, remember that for the 600k indels mentioned in the study their minimum length is 16bp, not their mean length, which means that it's likely that the overall indel contribution is far higher.

2:1 is still relatively low compared to the chimp/human divergence ratio of ~4:1. But it's very different from the 1:10 that you've been trying to show and it completely invalidates this particular line of argument against human evolution.

 

[sfs]

So, to summarize what we have so far:

1. The team searched out all indels with length more than (not sure about "and equal to") 16bp.
2. They found 606,093 such indels.

You've gone off the track here. They started with an algorithm that could detect indels 16 bp long or shorter, and then added a second algorithm to find longer ones. The total that they found from both methods was 600,000.

The group's second paper (the Genome Research paper I cited previously) goes into more detail about their results, and also represents a somewhat more refined analysis, I believe. There they list the number of indels they found of different types. The total is now 415,000, since they have screened out duplicates. The numbers are as follows (with lengths added by me in parentheses):

single base (1): 120938
repeat expansions of different lengths
repeat length 1 (1): 77047
repeat length 2 (2): 29460
repeat length 3 (3): 5454
repeat length 4 (4): 7456
repeat length 5 (5): 1456
repeat length 6 (6): 558
repeat length 7 (7): 293
repeat length 8 (8): 325
repeat length 9 (9): 205
repeat length 10 (10): 204
transposons (750): 2433
other (7.7): 169,607

I've used 750 as a typical transposon length (they quote the range 500 - 1000). To find the average length of the "other" category, you have to download supplemental table 1; the "other len" page has a breakdown by length. They range from length 2 to 9076. (Note that they all have to be at least 2 bp long, since the single-base indels were already listed in the table.)

These numbers give an overall average indel size of 8.3 bp. Note that this is the average for indels that are smaller than 10,000 bp, since that was the maximum they could detect in this study. They simultaneously identified SNPs, and found indels were 16.2% as common as SNPs. 0.162 x 8.3 = 1.34, i.e. indels contribute 1.34 times as many variable bases as SNPs in this study.

 

[shernren]

I gladly defer to the real geneticist.

The abstract of the paper sfs cited is available here:
http://www.genome.org/cgi/content/abstract/16/9/1182

and the supplementary data here:
http://www.genome.org/cgi/content/full/gr.4565806/DC1

The full text requires a paid subscription to access.

 

[mark kennedy]

I find it quite sad, myself.


Yes. And . . . what?


It's in fact quite ambiguous, and badly written. As the group's own numbers show, SNPs account for the majority of genetic variation as measured by events, not as measured by base pairs.

Now you are going to act as if a single nucleotide is ambiquise as a unit of measurement. Then you imply that a single nucleotide is just an event and could be more then one nucleotide. Then you make a semantical split between a single nucleotide and a base pair.

I wouldn't say this is sad, in fact it's actually tiresome.

10-20% of the base pair variation is caused by transposons. Transposons are only one half of one percent of all indels. How many base pairs do the other 99.5% of indels represent, Mark? Just answer that question.

The 1-2 million bases out of 10 million is proportion of indels to SNPs. If you have another 99% somewhere else I would love to hear about them because I know how this 99% could mean just about anything.

To answer it, you should look at the paper I quoted. It's from the same group that produced the transposon paper you're quoting from.

I assume this is what you are talking about:

In this report, we describe an initial map of human INDEL variation that contains 415,436 unique INDEL polymorphisms. These INDELs were identified with a computational approach using DNA re-sequencing traces that originally were generated for SNP discovery projects. They range from 1 bp to 9989 bp in length and are split almost equally between insertions and deletions, relative to the chimpanzee genome sequence. Five major classes of INDELs were identified, including (1) insertions and deletions of single-base pairs, (2) monomeric base pair expansions, (3) multi-base pair expansions of 2&#8211;15 bp repeat units, (4) transposon insertions, and (5) INDELs containing random DNA sequences. Our INDELs are distributed throughout the human genome with an average density of one INDEL per 7.2 kb of DNA.​

An initial map of insertion and deletion (INDEL) variation in the human genome

First they wrote a paper on the transposons, which contribute 10-20% as much as SNPs to base pair variation, and then they wrote a second paper about the other 99.5% of indels. The second paper makes it clear that indels taken as a whole contribute more base pairs of variation than do SNPs. Those are the facts. Interpret them however you wish, but stop getting the facts wrong.

So even though they are looking at indels that are 1/10 the size of SNPs, they are less then 1% of all the indels. Gotcha

Since transposons continue to be only a tiny fraction of all indels, your comment does not respond to my point.

Transposons are close to half the genome but a small fraction of the indels contributing to genetic variation. This would seem to be due to the fact that the transposons did their thing 35 million years ago and somehow stopped working. This reminds me of the enigmatic ERVs that became 8% of the mutation as the result of viruses making germline invasions.

It's not your point that I am having trouble with, it's the sweeping suppostion that is riddled throughout these discussions.

Transposons have been well measured. Long indels (which are much longer than transposons) have not been well measured. Therefore the total amount of base pairs introduced by indels has not been well measured. Even with the measurements that we do have, however, we can set a lower limit on it that is more than ten times higher than you claim is the case.

Farout! The lower limit is ten times greater then I have been saying, which means it's at least equal to the number of SNPs. They haven't measured the big indels very well so apparently they are only counting the little ones. Okay fine...

I don't give a fig whether you are convinced by this or by anything else.

I know that, you have that in common with every secularist I have encountered.

Believe that your head is a turnip, for all I care.

From now on I will consider my head a turnip with a cranial capacity three times that of chimpanzees. I still won't have a genetic basis for the unprecedented expansion so assuming it's a turnip makes very little difference to me.

I only respond to you so that readers won't be confused by the falsehoods you post.

So taking an explict statement at face value is now a lie? Tell me Steve, does the bp in 2 x 10^-9/bp/generation mean base pair yet or is that just insane?

My advice as a geneticist is for the casual reader to ignore anything and everything that Mark says about genetics.

My advice to anyone interested in genetics is to look at what they publish because they can't lie about it there.

 

[sfs]

The 1-2 million bases out of 10 million is proportion of indels to SNPs. If you have another 99% somewhere else I would love to hear about them because I know how this 99% could mean just about anything.

No, the 1-2 million bases is the proportion of bases due to transposons, not the proportion due to indels. It says so right in the part of the paper you keep quoting. Just reaad what you quote. The other 99% of indels are all of the classes I listed from their second paper (single-base indels, multimeric repeat expansions and "other"), which together occur 100 times as often.

So even though they are looking at indels that are 1/10 the size of SNPs, they are less then 1% of all the indels. Gotcha

They are looking at one small class of indels that are (summed up) 1/10 the size of SNPs. They are not (in this paper) looking at the other indels.

Transposons are close to half the genome but a small fraction of the indels contributing to genetic variation. This would seem to be due to the fact that the transposons did their thing 35 million years ago and somehow stopped working.

Mostly it's because transposons are fifty or 100 times as big as the average indel. But it is also true that there have been relatively few active transposons in recent human evolutionary history.

 

[mark kennedy]

I was going to wait until Mark does his "expositive review of six papers" argument and go through them one by one but since sfs jumped the gun I might as well jump in. If you take a careful look at the paper it's obvious that in terms of base pairs, indels outnumber SNPs in their contribution to human natural genetic variability.

I had hoped that there was a way to seperate the supposition from the actual science. I never really had a problem with someone coming to an informed opinion based on the actual evidence. The whole problem is that the measure of genetic variation is indicating that indels are about 1/10 of the human genetic variation. I was under the impression that some 10 million SNPs represented the the majoritiy of genetic variation between humans. Now there are these enigmatic indels being added to Hapmaps which should prove interesting over time.

First, the section actually cited by mark:

Together with previous studies, our analysis indicates that SNPs, indels, and transposon insertion polymorphisms represent significant sources of genetic variation in humans. Human populations are estimated to harbor &#8764;10 million common SNPs (Judson et al. 2002), &#8764;2 million common indels (our unpublished data), and &#8764;2000 common transposon insertion polymorphisms (this study). Therefore, with 10 million bases of variation, SNPs account for the majority of common human genetic variation, followed by indels and then transposon insertion polymorphisms. On the other hand, if we assume that the average transposon polymorphism in humans is &#8764;500&#8211;1000 bp in length, then the total amount of variation caused by common transposon insertions is 1&#8211;2 million base pairs (equivalent to 10&#8211;20% of the base pair variation caused by SNPs). Thus, in terms of the number of base pairs, common transposon insertions cause significant levels of human genetic variation. Moreover, humans also are likely to harbor >10 million rare private transposon insertions (cases in which only one or a few individuals have the insertion). Therefore, transposon insertion polymorphisms cause significant levels of human variation.​

(emphasis added)

Ok, you were driving this home to a point you wanted to make.

Pay attention to the bolded phrase: it signals that the argument has changed. "In terms of the number of base pairs", common transposon insertions cause significant levels of human genetic variation - even though SNPs outnumber transposon insertions roughly 5,000 to 1. Clearly this means that the preceding analysis ("SNPs account for the majority of common human genetic variation" etc.) was not in terms of the number of base pairs but in terms of the number of events.

Somehow the measure of base pairs as the unit of measurement has to be dismissed. There may well be a rational reason for this and we will see what that is, hopefully before I get started on the next thread.

This is semantics, mark, so criticize it as semantics if you will but there's more. This requires a close look at the methodology section (in which I get lost too). But first, a look at the abstract:

We began by identifying 606,093 insertion and deletion (indel) polymorphisms in the genomes of diverse humans. We then screened these polymorphisms to detect indels that were caused by de novo transposon insertions.​

"606,093 indel polymorphisms", aight? Just how long were those indels?

After the traces were successfully mapped to unique genomic locations, they were unmasked and aligned to their assigned genomic locations using the Bl2Seq program (NCBI). The Bl2Seq program allowed for as much as a 16-base gap in the alignments and led to identification of indels as large as 16 bases in length.​

A new algorithm also was developed to identify indels that were >16 bp in length. Our strategy was designed to split trace data into two blocks upon encountering a region in the pairwise alignment that no longer matched the query. The first block of sequence that matched was maintained in the correct position, and the nonmatching sequence was moved over as a block, 1 base at a time, until a match was obtained. The Perl program that was developed to accomplish this task moved the nonmatching block until it detected either a perfect alignment or a distance of 10,000 bases (the maximum distance allowed by the program). The 5 bases on each side of an indel candidate were required to have Phred scores of &#8805;20 to ensure that high-quality bases were being used to locate the indel junctions. Indel candidates were deposited into dbSNP under accession nos. ss8029278&#8211;ss8176133, ss8475737&#8211;ss8484870, ss14926095&#8211;ss15354938, and ss15357378&#8211;ss15378640.​

I took the liberty of bolding the references to base pairs as a unit of measurement for my own reference. It's just for clarification later.

Now, I'm a bit fuzzy on the Bl2Seq program (I'm an armchair geneticist ) but clearly their settings enabled them to identify indels with length 16 base pairs and above. The candidate indels were further winnowed with the Perl program mentioned in the second paragraph to find genuine and extremely convincing indels - this means that conceivably there were more indels longer than 16bp than detected. The indel sequences were catalogued, and if you look at the accession numbers you will note that 606,093 entries were added to the database - matching the number of indels noted in the abstract.

So, to summarize what we have so far:

1. The team searched out all indels with length more than (not sure about "and equal to") 16bp.
2. They found 606,093 such indels.

What does this tell us? 606,093 indels x 16 bp/indel gives us a minimum of 9.7 million base pairs. In other words, indels contribute at least 9.7 million base pairs of variation to the human genome. The ~10 million SNPs only contribute ~10 million base pairs, so this leaves indels and SNPs almost neck-to-neck.


Is this over? No way. Jumping down to the conclusion:

Together with previous studies, our analysis indicates that SNPs, indels, and transposon insertion polymorphisms represent significant sources of genetic variation in humans. Human populations are estimated to harbor &#8764;10 million common SNPs (Judson et al. 2002), &#8764;2 million common indels (our unpublished data), and &#8764;2000 common transposon insertion polymorphisms (this study). Therefore, with 10 million bases of variation, SNPs account for the majority of common human genetic variation, followed by indels and then transposon insertion polymorphisms.​

Note ~2 million common indels, out of which this study publishes the identification of roughly 600k. This means that there are still 1.4 million common indels with length < 16bp. And that's a lot. Assuming that the mean length of these indels is 8bp, the 1.4 million indels add up to 11.2 million base pairs - meaning that indels, giving in total 20.9 million base pairs, outnumber SNPs' 10 million base pairs in terms of base pairs 2 to 1.

And 20.9 million base pairs is something like an absolute minimum, remember that for the 600k indels mentioned in the study their minimum length is 16bp, not their mean length, which means that it's likely that the overall indel contribution is far higher.

2:1 is still relatively low compared to the chimp/human divergence ratio of ~4:1. But it's very different from the 1:10 that you've been trying to show and it completely invalidates this particular line of argument against human evolution.

Ok, so there are 10 million SNPs in a genome 3 billion base pairs long. That comes to just over 3% and then you want to start adding in the indels. Good luck with that.

 

[sfs]

So taking an explict statement at face value is now a lie?

I didn't say you were telling lies. I said you were repeating falsehoods. I have no doubt that you think your misunderstandings are true. Just because it's an explicit statement doesn't mean you can't misunderstand it. Take the following:

Tell me Steve, does the bp in 2 x 10^-9/bp/generation mean base pair yet or is that just insane?

Of course bp means base pair; I have never suggested otherwise. The misunderstanding on your part is in thinking that this means that mutations are measured in base pairs. They're not. The sequence that's doing the mutating is measured in base pairs.

The mutation rate is 2x10^-8 mutations per base pair per generation. That is exactly the same as saying that the mutation rate is 20 mutations per billion base pairs per generation. Since the human genome has 3 billion base pairs, the mutation rate for humans is therefore 60 mutations per genome copy per generation. Do you think that mutations are measured in genome copies?

 

[mark kennedy]

No, the 1-2 million bases is the proportion of bases due to transposons, not the proportion due to indels. It says so right in the part of the paper you keep quoting. Just reaad what you quote. The other 99% of indels are all of the classes I listed from their second paper (single-base indels, multimeric repeat expansions and "other"), which together occur 100 times as often.

Ok, here's that section:

Together with previous studies, our analysis indicates that SNPs, indels, and transposon insertion polymorphisms represent significant sources of genetic variation in humans. Human populations are estimated to harbor &#8764;10 million common SNPs, &#8764;2 million common indels (our unpublished data), and &#8764;2000 common transposon insertion polymorphisms (this study). Therefore, with 10 million bases of variation, SNPs account for the majority of common human genetic variation, followed by indels and then transposon insertion polymorphisms. On the other hand, if we assume that the average transposon polymorphism in humans is &#8764;500&#8211;1000 bp in length, then the total amount of variation caused by common transposon insertions is 1&#8211;2 million base pairs (equivalent to 10&#8211;20% of the base pair variation caused by SNPs). Thus, in terms of the number of base pairs, common transposon insertions cause significant levels of human genetic variation. Moreover, humans also are likely to harbor >10 million rare private transposon insertions (cases in which only one or a few individuals have the insertion). Therefore, transposon insertion polymorphisms cause significant levels of human variation.​

They are looking at one small class of indels that are (summed up) 1/10 the size of SNPs. They are not (in this paper) looking at the other indels.

I don't know anything about the other transposons, they said this though:

Moreover, humans also are likely to harbor >10 million rare private transposon insertions (cases in which only one or a few individuals have the insertion).​

Mostly it's because transposons are fifty or 100 times as big as the average indel. But it is also true that there have been relatively few active transposons in recent human evolutionary history.

Supposedly transposons did their thing about 35 million years ago. As memory serves that was about the time ERVs were doing their thing but that's ancient history, prehistoric even.

I'm itching to get on to the new thread, correct my errors at will in this one Steve.

 

[Mallon]

Given Mark's arguments here and his recent tirade about evolution in schools (see Creationist subforum), I have to ask: Mark, what is your formal education and training as far as genetics and evolution go?

 

[Deamiter]

I had hoped that there was a way to seperate the supposition from the actual science. I never really had a problem with someone coming to an informed opinion based on the actual evidence. The whole problem is that the measure of genetic variation is indicating that indels are about 1/10 of the human genetic variation. I was under the impression that some 10 million SNPs represented the the majoritiy of genetic variation between humans. Now there are these enigmatic indels being added to Hapmaps which should prove interesting over time.

Your impression is wrong. Well, sort of. sfs just explained HOW this is wrong, but you went on to repeat yourself (maybe you hadn't gotten to that post yet). I just want to emphesize this because I keep seeing you say this over and over. SNPs are the vast majority of mutations (events) and a minority in terms of differing base pairs. When they use the term "genetic variation" they're talking about EVENTS here. sfs (the resident geneticist) says this is poorly written, so I assume it's not some sort of field standard or anything, just a poor choice of words.

Since they gave units for each of their numbers, however, there's no reason to misunderstand this basic point based on this poorly written sentence!

I think sfs said it better in post 43:

It's in fact quite ambiguous, and badly written. As the group's own numbers show, SNPs account for the majority of genetic variation as measured by events, not as measured by base pairs.

And what's this obsession with the percentage of the genome? Everybody agrees that the magazine article you keep bringing up was poorly written in that they didn't give units to their 99%. You keep contrasting this 99% with the difference in terms of base pairs, but nobody here argues that the base pairs in chimps and humans ARE 99% the same. I doubt the magazine article you so hate said that either, but it's impossible to tell because they didn't give units.

I mean, you can keep bringing it up if it makes you feel better, but nobody's claiming that there's less than 1% diveregence in terms of base pairs, so it's a bit of a straw man if you're trying to show that somebody on this board is wrong about something.

 

[mark kennedy]

Given Mark's arguments here and his recent tirade about evolution in schools (see Creationist subforum), I have to ask: Mark, what is your formal education and training as far as genetics and evolution go?

Not much, I took a Biology 101 course and never used the textbook in the class. I thought it odd that evolution was ever an issue and I'm finally coming to the general conclusion that it never was.

I meant to put that in the Origins common area but I guess I forgot. At any rate, to answer your question, not much.

 

[busterdog]

Given Mark's arguments here and his recent tirade about evolution in schools (see Creationist subforum), I have to ask: Mark, what is your formal education and training as far as genetics and evolution go?

To paraphrase Dave Barry. Let's say that you wanted to build a super-conducting super-collider. You go to the Department of Marge, who drives a bus or something. And you proposed, Marge, we want to spend 50 billion dollars to take things we can't see, make them go faster than we have ever made anything go just to smash them up and see what comes out of it. Marge says, I don't think so.

If you can't get it past us, maybe you are not doing something right.

However, I am quite proud of my 99 on the New York State biology regents in 1978.

 

[mark kennedy]

Your impression is wrong. Well, sort of. sfs just explained HOW this is wrong, but you went on to repeat yourself (maybe you hadn't gotten to that post yet). I just want to emphesize this because I keep seeing you say this over and over. SNPs are the vast majority of mutations (events) and a minority in terms of differing base pairs. When they use the term "genetic variation" they're talking about EVENTS here. sfs (the resident geneticist) says this is poorly written, so I assume it's not some sort of field standard or anything, just a poor choice of words.

We seem to be butting heads over the measure of mutations. Most of the scientific literature seems to focus on either nucleotides or base pairs. It's usually something like 2 x 10^-8 which is 2 per 100 million base pairs which comes to about 60 base pairs for every human zygote.

I am getting the general impression that when considering those 60 base pairs I am supposed to accept that it does not count the indels or that one base pair is actually a larger mutation. That makes absolutely no sense to me.

Since they gave units for each of their numbers, however, there's no reason to misunderstand this basic point based on this poorly written sentence!

The sentence was not poorly written enough to account for the measure being in base pairs, there is no basis for this in the paper or any of the scientific literature.

I think sfs said it better in post 43:

It's in fact quite ambiguous, and badly written. As the group's own numbers show, SNPs account for the majority of genetic variation as measured by events, not as measured by base pairs

Notice the use of base pairs again and again:

They are measured in terms of base pairs and the percentage of genome represented here in Table 1.

Transposon insertion polymorphisms identified in humans

Notice the size of the indels is measured in base pairs in Table 5.

Insertion polymorphisms generated by other forms of mobilized DNA

Now for the poorly written statement that comes right out and makes the comparison:

Therefore, with 10 million bases of variation, SNPs account for the majority of common human genetic variation, followed by indels and then transposon insertion polymorphisms. On the other hand, if we assume that the average transposon polymorphism in humans is ~500&#8211;1000 bp in length, then the total amount of variation caused by common transposon insertions is 1&#8211;2 million base pairs (equivalent to 10&#8211;20% of the base pair variation caused by SNPs).​

There is a reason that the actual base pairs involved is critical. In the comparison of the Chimpanzee genome and the human genome the indels dwarfed the size of the single nucleotide insertions. In this comparison it was quite the opposite. I'm being told that I just misunderstood but I'm still not buying it.

And what's this obsession with the percentage of the genome? Everybody agrees that the magazine article you keep bringing up was poorly written in that they didn't give units to their 99%. You keep contrasting this 99% with the difference in terms of base pairs, but nobody here argues that the base pairs in chimps and humans ARE 99% the same. I doubt the magazine article you so hate said that either, but it's impossible to tell because they didn't give units.

The fact is that Time and Nature both said that the Chimpanzee and Human genomes had 98% the same DNA. This is not true and represents a difference of over 100 million base pairs. Mutation rates are almost allways given in terms of base pairs, I have yet to see an exception.

"RATES of spontaneous mutation per replication per measured target vary by many orders of magnitude depending on the mutational target size (from 1 to >1010 b, where b stands for base or base pair as appropriate)

...With 6.4 x 10^9 base pairs in the diploid genome, a mutation rate of 10^-8 means that a zygote has 64 new mutations. It is hard to image that so many new deleterious mutations each generation is compatible with life, even with an efficient mechanism for mutation removal. Thus, the great majority of mutations in the noncoding DNA must be neutral."

Table 5. Mutation rates estimated from specific loci in higher eukaryotes


Rates of Spontaneous Mutation


Now I would like you to do me the courtesy of considering this common measure of mutations as applied to the known divergance between chimpanzees and humans. Then I would like for you to give yourself a minute and think about 125 million base pairs in 5 million years because it comes to 25 per year for 5 million years or 500 base pairs per generation (estimated at 20 years).

This simply does not happen and what sfs is trying to do is convince everyone that the 90 million base pairs are the same as 5 million mutation events. If you do that then the formula I just gave you is absolutly meaningless.

I mean, you can keep bringing it up if it makes you feel better, but nobody's claiming that there's less than 1% diveregence in terms of base pairs, so it's a bit of a straw man if you're trying to show that somebody on this board is wrong about something.

Something is very wrong and it is not the board. The mutation rate applied to human evolution does not account for the divergance between chimpanzees and humans. It was all cut and dried when it was 98% but now estimates are around 95% and creationists are saying it is more like 90%.

The postion of evolutionists is completly indefensible unless they can make every indel count as one base pair which is exactly what they are trying to do.

 

[Deamiter]

Now I would like you to do me the courtesy of considering this common measure of mutations as applied to the known divergance between chimpanzees and humans. Then I would like for you to give yourself a minute and think about 125 million base pairs in 5 million years because it comes to 25 per year for 5 million years or 500 base pairs per generation (estimated at 20 years).

This simply does not happen and what sfs is trying to do is convince everyone that the 90 million base pairs are the same as 5 million mutation events. If you do that then the formula I just gave you is absolutly meaningless.

If the average bp per mutation is 9 (didn't we calculate about 7.7 WITHOUT the largest?) then the average number of base pairs could easily be 90M with a NUMBER of mutations of 5M. What about this doesn't make sense?

As for the calculation of 500 base pairs per generation... whatever is wrong with that? If you've got a population of 10,000 that's 1/20 per member. Since we're only counting FIXED mutations, that seems entirely reasonable to me. Certainly within an order of magnitude or so.

 

[Deamiter]

Now for the poorly written statement that comes right out and makes the comparison:

Therefore, with 10 million bases of variation, SNPs account for the majority of common human genetic variation, followed by indels and then transposon insertion polymorphisms. On the other hand, if we assume that the average transposon polymorphism in humans is ~500&#8211;1000 bp in length, then the total amount of variation caused by common transposon insertions is 1&#8211;2 million base pairs (equivalent to 10&#8211;20% of the base pair variation caused by SNPs).

There is a reason that the actual base pairs involved is critical. In the comparison of the Chimpanzee genome and the human genome the indels dwarfed the size of the single nucleotide insertions. In this comparison it was quite the opposite. I'm being told that I just misunderstood but I'm still not buying it.

did you miss sfs repeatedly reminding you that transposons are a small fraction of all indels? He noted this in post 43 and 48 and at least one other place somewhere between there and here. Why do you keep claiming that this quote says that indels are more prominent (in terms of base pairs) when this quoted paper CLEARLY only addresses transposons (a small subset of indels)?

 

[shernren]

Mutation rates are almost allways given in terms of base pairs, I have yet to see an exception.

"RATES of spontaneous mutation per replication per measured target vary by many orders of magnitude depending on the mutational target size (from 1 to >1010 b, where b stands for base or base pair as appropriate)

...With 6.4 x 10^9 base pairs in the diploid genome, a mutation rate of 10^-8 means that a zygote has 64 new mutations. It is hard to image that so many new deleterious mutations each generation is compatible with life, even with an efficient mechanism for mutation removal. Thus, the great majority of mutations in the noncoding DNA must be neutral."

Table 5. Mutation rates estimated from specific loci in higher eukaryotes


Rates of Spontaneous Mutation

Uh, mark, do you read what you cite? The abstract reads (formatted for clarity) :

Rates of spontaneous mutation per genome as measured in the laboratory are remarkably similar within broad groups of organisms but differ strikingly among groups.
Mutation rates in RNA viruses, whose genomes contain ca. 104 bases, are roughly 1 per genome per replication for lytic viruses and roughly 0.1 per genome per replication for retroviruses and a retrotransposon.
Mutation rates in microbes with DNA-based chromosomes are close to 1/300 per genome per replication; in this group, therefore, rates per base pair vary inversely and hugely as genome sizes vary from 6 x 103 to 4 x 107 bases or base pairs.
Mutation rates in higher eukaryotes are roughly 0.1&#8211;100 per genome per sexual generation but are currently indistinguishable from 1/300 per cell division per effective genome (which excludes the fraction of the genome in which most mutations are neutral). It is now possible to specify some of the evolutionary forces that shape these diverse mutation rates.

Look at the units used for mutation rates (underlined) :

1 per genome per replication
0.1 per genome per replication
1/300 per genome per replication
0.1-100 per genome per sexual generation
1/300 per cell division per effective genome

None of these units for mutation rates involve base pairs in any way. The only things described in terms of base pairs (italicized) are genome sizes and mutation lengths, not mutation rates or number of mutations.

Exceptions, right there.

And here's a preemptive discussion:

[SIZE=-1]Homo sapiens:[/SIZE] The human data are less reliable than the C. elegans, Drosophila and mouse data. A number of dominant-mutation rates have been inferred from the frequency of affected children of normal parents, and sometimes confirmed by equilibrium estimates for those dominants with severe effects. These values range from 10-4 to 10-6, with a rough average of 10-5 (V[SIZE=-1]OGEL[/SIZE] and M[SIZE=-1]OTULSKY[/SIZE] 1997). For genes of size 103 b, this corresponds to a rate of 10-8 per b per generation. An estimate based on specific changes in the hemoglobin molecule gave 0.74 x 10-8 per b per generation (V[SIZE=-1]OGEL[/SIZE] and M[SIZE=-1]OTULSKY[/SIZE] 1997), but this is clearly an underestimate because other kinds of changes are not included. A third, quite independent estimate is based on rates of evolution of pseudogenes in human ancestry, which are likely to be identical to mutation rates (K[SIZE=-1]IMURA[/SIZE] 1983[SIZE=-1]A[/SIZE]). This gives about 2 x 10-8 per b per generation (C[SIZE=-1]ROW[/SIZE] 1993, C[SIZE=-1]ROW[/SIZE] 1995). We shall take 10-8 as a representative value. However, because the overwhelming majority of human mutations occur in males (see below), the male rate must be about twice the average rate, or 2 x 10-8. The number of cell divisions prior to sperm formation in a male of age 30 is about 400 (D[SIZE=-1]ROST[/SIZE] and L[SIZE=-1]EE[/SIZE] 1995; V[SIZE=-1]OGEL[/SIZE] and M[SIZE=-1]OTULSKY[/SIZE] 1997). Thus, µb

2 x

= 5 x 10-11. For 8 x 107 genes (B[SIZE=-1]IRD[/SIZE] 1995) of average size 103 b, µeg

0.004 and µegs

1.6.

An alternative method for estimating µegs has been proposed by K[SIZE=-1]ONDRASHOV[/SIZE] and C[SIZE=-1]ROW[/SIZE] 1993 based on the idea that purely neutral sequences such as pseudogenes can be used as a benchmark to identify sites which show clear evidence of selective constraints. If the abundance of such sites can be determined in this way, the effective genome size and its mutation rate can be estimated purely from rates of DNA sequence evolution. This method has yet to be applied to large quantities of sequence data. For hemoglobin genes, about 15% of bases seem to be under the effective control of selection, which may be about average for genes encoding proteins; for a more sophisticated treatment, see K[SIZE=-1]IMURA[/SIZE] 1983[SIZE=-1]B[/SIZE].
With 6.4 x 109 base pairs in the diploid genome, a mutation rate of 10-8 means that a zygote has 64 new mutations. It is hard to image that so many new deleterious mutations each generation is compatible with life, even with an efficient mechanism for mutation removal. Thus, the great majority of mutations in the noncoding DNA must be neutral.​

Unit used: 2 x 10^-8 per basepair per generation. Keep your eyes on that while I trot out an analogy.

Let's say the accident rates in KL are 2 x 10^-8 per car per day. Now, if there are ten million cars in KL on any given day, and there are 365 days in a year, can I then say that only 73 cars will be affected by accidents every year? That's not true. What I know is that 73 accidents will happen every year:

(2 x 10^-8 / car / day) x (10 million cars) x (365 days)
= 73

The resultant "73" is unitless, but because we are told beforehand that accidents are being measured, we know that 73 should be the number of accidents. Does it follow that 73 cars are affected? Not necessarily: accidents can affect more than one car at a time. If a 25-car pileup happened on New Year's Day I wouldn't say "Ok, that's 25 accidents, 48 more to go in the year!" An accident would still be one accident no matter how many cars had been involved in it.

In the same way, their calculation is:

No of mutations =
10 ^ -8 per base pair per generation x 6.4 x 10^9 base pairs x 1 generation
= 64.

Does this "64" come with a unit? No, the units cancel each other out to give a unitless quantity as in the example above. However, because it is explicitly noted that the things-being-counted in this case are mutations, it is obvious that these are 64 mutations instead of 64, say, base pairs (in which case the unit should come out as "base pairs" in the above calculation).

I hope this settles it once and for all ...

 

[mark kennedy]

Uh, mark, do you read what you cite? The abstract reads (formatted for clarity) :

Actually I am very familar with it but apparently there is no such thing as a well read creationist. That's bases per genome per generation, that all it is, it's a formula and what you are leading up to is non sequitor at best.

Look at the units used for mutation rates (underlined) :

1 per genome per replication

1 what per what number of base pairs=genome?

0.1 per genome per replication

0.1 what, per what number of base pair in the genome?

1/300 per genome per replication

1/300 what per what number of base pairs in the genome?

0.1-100 per genome per sexual generation

0.1-100 what, per what number of base pairs in the genome?

1/300 per cell division per effective genome

1/300 what, what cell is dividing, how many base pairs are effective in the genome?

You make this huge effort to appear as if you are being precise and systematic when you are argueing in circles around the main point. The mutations are measured in base pairs and don't you think by now I know that I have to know my material cold? No creationist could get away with the flagerant misrepresentation you just posted here.

None of these units for mutation rates involve base pairs in any way.

I want you to think about this because frankly, I am sick and tired of being told this when I know it's not true:

Mutation rates in RNA viruses, whose genomes contain ca. 104 bases, are roughly 1 per genome per replication​

That's 1 base in the virus which is 104 bases long.

Mutation rates in microbes with DNA-based chromosomes are close to 1/300 per genome per replication; in this group, therefore, rates per base pair vary inversely and hugely as genome sizes vary from 6 x 103 to 4 x 107 bases or base pairs.​

The rates per genome varies here because the genome sizes vary greatly. It is from 6 x 103 to 4 x 107 bases or base pairs.

Mutation rates in higher eukaryotes are roughly 0.1/100 per genome per sexual generation but are currently indistinguishable from 1/300 per cell division per effective genome

The size of the effective genome is a very small fraction of the genome in eukaryotes (you are composed of eukaryotes by the way). Dispite that fact the per genome/per sexual generation rate is indisitiquishable.

RATES of spontaneous mutation per replication per measured target vary by many orders of magnitude depending on the mutational target size (from 1 to >1010 b, where b stands for base or base pair as appropriate),

The rate varies because the size of the target varies from 1 to greator then 1010. The b stand for base or base pair.

the average mutability per b (from 10-4 to 10-11 per b per replication),

The mutability per base, do you understand that this is per base!!!???

and the specific mutability of a particular b (which can vary by >10^4-fold).

That means that the mutability of a base can vary by a10,000 fold magnitude.

The only things described in terms of base pairs (italicized) are genome sizes and mutation lengths, not mutation rates or number of mutations.

Horse feathers!!! Everything is measured in base pairs, the length of the section being viewed, the length of the mutation and the length of the respective genomes.

Exceptions, right there.

The quoted portion <snip> is in base pairs, everything in the paper is in base pairs, everything in every genome in every living system known to man is in base pairs.

Keep your eyes on that while I trot out an analogy.

Dude I love you but I have got to be honest here, you have bought and are trying to sell a lie. I really don't have a problem one with TEs, most of them express genuine Christian conviction and I enjoy exchanges with them. You have been told wrong and apparently bought it hook line and sinker. I know that you believe what you are saying, I can tell by your attitude. My problem is not with you, it's with the one who told the original falsehood.

Let's say the accident rates in KL are 2 x 10^-8 per car per day.

Stop right there, here you are looking at 2 x 10^-8 which is 2 times a decimal point followed by 9 zeros. I am no statistician but 2 cars having accidents out of 100 million cars is pretty darn good.

Now, if there are ten million cars in KL on any given day, and there are 365 days in a year, can I then say that only 73 cars will be affected by accidents every year? That's not true. What I know is that 73 accidents will happen every year:

(2 x 10^-8 / car / day) x (10 million cars) x (365 days)
= 73

You are measureing you rate per day, you have 2 cars out of 10^8 or 1,000,000,000 cars.

The resultant "73" is unitless,

There were 73 cars involved in accidents that year, which by the way is some good driving for the number of cars on the road there.

I can't let you continue with the analogy, it's just too embarasing.

In the same way, their calculation is:

No of mutations =
10 ^ -8 per base pair per generation x 6.4 x 10^9 base pairs x 1 generation
= 64.

This is what I have come to dispise about evolution, this is how comprehensive formulas turn into mindless gibberish.

With 6.4 x 10^9 base pairs in the diploid genome, a mutation rate of 10^-8 means that a zygote has 64 new mutations. It is hard to image that so many new deleterious mutations each generation is compatible with life, even with an efficient mechanism for mutation removal. Thus, the great majority of mutations in the noncoding DNA must be neutral.​

The genome is 3.2 billion base pairs long, or thought to be in 1998 when this paper was published. A diploid generation means that there are two copies of the genome during meiosis. The 10^-8 means 1 per 100,000,000 bases so out of 1 billion bases there will be ten mutations. Out of 3 billion there will be 30 and in the diploid generation there will be approximatly 60.