New genes

[Steve660]

Lets just take the DNA difference. For a long time we we told that the 1% to 1.5% difference was good evidence that we are related to chimps. Now we find that the difference is nearer 30% and no-one bats an eyelid.

What's your source for this claim? I suspect you've got the numbers muddled. Or some other creationist has and you've just repeated the error. The numbers may be referring to two separate things. In terms of base pairs, human & chimp DNA are indeed about 1.5 % different. Not 30 %. In terms of genes the number will be much higher, maybe 30 %, because it takes only one base pair to be different for the gene to be counted as different. Perhaps it should be that 30 % of the genes are different, and they differ by about 1.5 % of their base pairs. Unless you define what exactly each number refers to the argument is highly misleading. Like so many creationist arguments.

[marcsurtees]

Lets just take the DNA difference. For a long time we we told that the 1% to 1.5% difference was good evidence that we are related to chimps. Now we find that the difference is nearer 30% and no-one bats an eyelid.

What's your source for this claim?

It just so happens that I was preparing a slide on this for my next debate, here it is:
Non-alignable(1) ~24%
Single Nucleotide Polymorphism(1) ~3%(of 76% that is alignable)
Insertion/deletion(1) 1.2%(of 76%)
Total difference about 30%

1Nature 437:69-87
2Nature 437:88-93
3PLoS ONE 1:e85

[Brian Jordan]

It just so happens that I was preparing a slide on this for my next debate, here it is:
Non-alignable(1) ~24%
Single Nucleotide Polymorphism(1) ~3%(of 76% that is alignable)
Insertion/deletion(1) 1.2%(of 76%)
Total difference about 30%

And these changes took place when? 1 million years ago, 6000 years ago or in the '50s on the back row of the Roxy while watching King Kong?

[marcsurtees]

It just so happens that I was preparing a slide on this for my next debate, here it is:
Non-alignable(1) ~24%
Single Nucleotide Polymorphism(1) ~3%(of 76% that is alignable)
Insertion/deletion(1) 1.2%(of 76%)
Total difference about 30%

And these changes took place when? 1 million years ago, 6000 years ago or in the '50s on the back row of the Roxy while watching King Kong?

This is a question that is not possible to answer with the currently available data, and rather tends to obscure the significance of the facts.

The important point is that the human and chimp genomes are approx 30% different and this is consistent with humans and chimps being different, which is a prediction of special creation.

[Chris Sergeant]

humans and chimps being different, which is a prediction of special creation.

How profound. Didn’t realise it took special creation to work that out.

Non-alignable(1) ~24%

Should that be ~24% not available for alignment, because the chimpanzee genome was not fully sequenced at the time of the study in question (2005).
You approach is like reading 76% of two books and finding them virtually identical. Then deducing that the remaining 24% of pages must be completely different, because you have not read them.

http://efference.blogspot.com/2008/11/h ... h-and.html

[Brian Jordan]

this is consistent with humans and chimps being different, which is a prediction of special creation.

Prediction? Come off it, creationism doesn't predict anything it declares things. Then it bends and distorts to fit things into its predetermined mould.

[marcsurtees]

Should that be ~24% not available for alignment, because the chimpanzee genome was not fully sequenced at the time of the study in question (2005).
You approach is like reading 76% of two books and finding them virtually identical. Then deducing that the remaining 24% of pages must be completely different, because you have not read them.

http://efference.blogspot.com/2008/11/h ... h-and.html

Hi Chris,

No it should not. If you had read the original paper that I cited you would know that 94% of the chimp genome was sequenced. The authors then aligned all that they could and found that
"Best reciprocal nucleotide-level alignments of the chimpanzee and human genomes cover ~ 2.4 gigabases (Gb) of high-quality sequence, including 89 Mb from chromosome X and 73 Mb from chromosome Y."

That makes 2.4 of 3.2 (the number of gigabases in the human genome) which is 75%.

It's always a good idea to check the original source, rather than take the word of a blogger who apparently did not.

[Steve660]

It just so happens that I was preparing a slide on this for my next debate, here it is:
Non-alignable(1) ~24%
Single Nucleotide Polymorphism(1) ~3%(of 76% that is alignable)
Insertion/deletion(1) 1.2%(of 76%)
Total difference about 30%

1Nature 437:69-87
2Nature 437:88-93
3PLoS ONE 1:e85

I have been checking these references, or at least trying to. The last has so far eluded me as insufficient information is provided for the last one. The PLoS ONE site seems not to lend itself to searching by volume or page numbers. Please provide more information, or (better) a link.

The second deals with duplications, and concludes with:
"Nevertheless, when compared to single-base-pair differences, which
account for 1.2% genetic difference, base per base, large segmental
duplication events have had a greater impact (2.7%) in altering the
genomic landscape of these two species."

Note this is on a base per base basis. As one duplication will duplicate many bases, it will have a disproportionate effect. It would be more reasonable to count a duplication as one difference, on a par with a single base pair change.

A similar argument applies to indels, which again affect multiple base pairs. They should, therefore, be seen as one difference each, instead of multiple differences (multiple differences would mean each base pair difference being individually counted). The ca. 3 % of base pair differences caused by indels then becomes about 0.4 % difference, as indels are about 7 times less common than single base pair changes. Add this on to the 1.2 % difference due to single base pair changes and we have 1.6 % difference, plus the difference from duplication events, which as I explained above may be much, much less than 2.7 % if one counts duplications as single differences. An overall difference of 98 % may be quite reasonable.

I can find no mention of your "Non-alignable(1) ~24%" figure in either of these Nature papers. Please provide precise details - page number, column, paragraph. I'm not sure that this figure is meaningful anyway. It only takes one base pair difference to render a sequence of DNA non-alignable. It is a bit like declaring a whole page non-alignable with the corresponding one in another edition of a book because one single letter is missing, thereby throwing the whole sequence out by one. The authors also comment that "nearly all of the bases are identical by descent and sequences can be readily aligned except in recently derived, large repetitive regions."

The authors of these papers do not consider human and chimp DNA to be radically different, and certainly not by 30 %. Indeed they conclude with an affirmation of our striking similarities:
"We hope that elaborating how few differences separate our
species will broaden recognition of our duty to these extraordinary
primates that stand as our siblings in the family of life."

See also http://www.talkorigins.org/indexcc/CB/CB144.html for a brief summary of the importance of keeping to a consistent method of counting genetic differences, and that if you do this then chimps come out as our closest relatives. It looks to me like you are trying to fiddle the figures to ramp up the differences as much as you can. A standard creationist ploy of distorting the evidence to make it fit.

I think you will need to change your slide, or (better) ditch it altogether.

[marcsurtees]

It just so happens that I was preparing a slide on this for my next debate, here it is:
Non-alignable(1) ~24%
Single Nucleotide Polymorphism(1) ~3%(of 76% that is alignable)
Insertion/deletion(1) 1.2%(of 76%)
Total difference about 30%

1Nature 437:69-87
2Nature 437:88-93
3PLoS ONE 1:e85

I have been checking these references, or at least trying to. The last has so far eluded me as insufficient information is provided for the last one. The PLoS ONE site seems not to lend itself to searching by volume or page numbers. Please provide more information, or (better) a link.

I think you will need to change your slide, or (better) ditch it altogether.

Actually I cut and pasted some of the slide and I should have only given the first reference as the others support other points that I deleted as they were not directly related to the point I was trying to make.

The quote that I provided comes from page 71 and is the first sentence after the sub head "Nucleotide diveregence". This is direct evidence for the claim, that only 76% of the chimp genome alignes with the human genome.

All the other minor differences that you refer to are from the alignable portions which of course does not require absolute base by base comparision. In fact to be alignable the difference have to be small as shown. The authors actually ignore the 24% non alignable part and make comparisons on what is comparable.

[marcsurtees]

See also http://www.talkorigins.org/indexcc/CB/CB144.html for a brief summary of the importance of keeping to a consistent method of counting genetic differences, and that if you do this then chimps come out as our closest relatives.

I know some orangutans who would beg to differ!

See
http://www.livescience.com/animals/0906 ... tives.html

[psiloiordinary]

Hi Marc,

No. That article does not base any conclusions on a genetic analysis.

Cheers,

Psi

[Peter Henderson]

See also http://www.talkorigins.org/indexcc/CB/CB144.html for a brief summary of the importance of keeping to a consistent method of counting genetic differences, and that if you do this then chimps come out as our closest relatives.

I know some orangutans who would beg to differ!

See
http://www.livescience.com/animals/0906 ... tives.html

are you referring to the orangutans or the scientists Marc:

In other words, if the DNA evidence that many biologists use as evidence turned out not to accurately reveal evolutionary relationships, the work of many molecular biologists would be suspect.

"If this was true, we would lose entire departments at major universities," Disotell said. "I would have nothing to do. I would go become a carpenter."

[Steve660]

I know some orangutans who would beg to differ!

See
http://www.livescience.com/animals/0906 ... tives.html

You actually know these orangutans?

That makes 2.4 of 3.2 (the number of gigabases in the human genome) which is 75%.

Elsewhere you say 76% is alignable, so the 25% that allegedly was not becomes 24%. Is there a reason for 25% becoming 24%, or are you just being generous?
If your interpretation of this data is correct then I think the authors would have had something to say about it. Instead they say, "nearly all of the bases are identical by descent and sequences can be readily aligned except in recently derived, large repetitive regions". I wonder how much of the 25 (or 24) % is composed of those "recently derived large repetitive regions".

[sfs]

It just so happens that I was preparing a slide on this for my next debate, here it is:
Non-alignable(1) ~24%
Single Nucleotide Polymorphism(1) ~3%(of 76% that is alignable)
Insertion/deletion(1) 1.2%(of 76%)
Total difference about 30%

1Nature 437:69-87
2Nature 437:88-93
3PLoS ONE 1:e85

Someone let me know about this thread because I was one of the authors on the chimpanzee genome paper. I'm afraid your calculation is wrong. Your basic problem is that "non-alignable" does not mean "different". Here's how the numbers actually work:

For reciprocal alignment, what matters is the size of the genome that has been sequenced, not the size of the whole genome. Despite statements saying that we have completed sequencing the human genome, what we have actually sequenced is the euchromatic genome, which is the whole genome except for the centromeres, telomeres and heterochromatic regions. This amounts to about 2.85 Gb (based on the paper describing the finished human genome, Nature 431:931-945 (2004)). So the maximum amount of chimpanzee genome that could have been aligned to human was 2.85 Gb. The rest has never been looked at in detail in either chimps or humans.

The second factor is the 94% coverage of the chimpanzee genome; that was the fraction where those assembling the genome had enough reads to be able to say anything about the sequence. Therefore there must have been at most 2.68 Gb of chimpanzee genome that could have been aligned to human.

And then there is the 1.5% of chimpanzee DNA that the paper reports is not present in human (made up of both insertions in chimpanzee and deletions in huma), i.e. the fraction that is genuinely different. Removing that means that we should expect 2.64 Gb to have aligned.

Looking at supplemental note 2 for the chimpanzee genome paper, we can see that 2.41 Gb of chimpanzee actually did align to assembled human DNA, and that another 0.24 Gb could be aligned to specific human chromosomes, but could not be placed more accurately than that due to the incompleteness of the human assembly; i.e. the chimp DNA clearly matches human DNA, but we weren't sure exactly where the human sequence fit in our own genome. (This fraction would have been a lot smaller if a later human assembly had been used, but the work was done before the human genome was "finished".) Thus a total of 2.65 Gb of chimpanzee actually aligned, compared to the 2.64 Gb expected. (Which is well within the uncertainty, especially given the roughness of the estimate of ~94% coverage.)

So there is no discrepancy. I will also add that, had there actually been an unexpected, massive discrepancy, we would have been falling over each other to report the finding, issue press releases and try to hog as much of the limelight as possible (after we checked and rechecked that we hadn't made a mistake, of course -- nothing like reporting an exciting new discovery and then finding out it was just a bug in a grad student's program). Why on earth would we hide such a momentous discovery?

[Steve660]

Brilliant response from sfs. Checkmate Marc. Now ditch that slide.