Basic Math Challenge: Chimpanzee and Human DNA divergence

[mark kennedy]

I have a very simple question, it's not that hard if you are willing to do a little addition. This is from post #242, in the Dover Trial thread. Talk Origins made a simple error, they equivocated the number of insertion/deletion events required for the divergence between the Chimpanzee and Human genomes as measured in base pairs. Did they get the math right or wrong?

I have issues with the Darwinian philosophy of natural history for one reason, the Scriptures are clear, God created life. If you are anyone else is convinced that Darwinian evolution has made it's case conclusively I say go in peace I have no problem with you. I'm just not going to pretend what they are telling me about the actual scientific evidence is true when I know for a fact it's otherwise. This is what I'm talking about, a statement that is corrected and easily refuted with basic math:

The difference between chimpanzees and humans due to single-nucleotide substitutions averages 1.23 percent, of which 1.06 percent or less is due to fixed divergence, and the rest being a result of polymorphism within chimp populations and within human populations. Insertion and deletion (indel) events account for another approximately 3 percent difference between chimp and human sequences, but each indel typically involves multiple nucleotides. The number of genetic changes from indels is a fraction of the number of single-nucleotide substitutions (roughly 5 million compared with roughly 35 million). So describing humans and chimpanzees as 98 to 99 percent identical is entirely appropriate (Chimpanzee Sequencing 2005). (Talk Origins, Claim CB144)
​

The question is what is 1.23% plus 3%, this isn't a trick question, it's not between 1% and 2% it's 4.23%. That's not my opinion, that's not my interpretation, that's exactly what the Initial Sequence of the Chimpanzee Genome paper, that they specifically cite, actually says:

Genetic differences that have accumulated since the human and chimpanzee species diverged from our common ancestor, constituting approximately thirty-five million single-nucleotide changes, five million insertion/deletion events,

• Single-nucleotide substitutions occur at a mean rate of 1.23%
• we estimate that the human and chimpanzee genomes each contain 40–45 Mb
• the indel differences between the genomes thus total ~90 Mb.

This difference corresponds to ~3% of both genomes and dwarfs the 1.23% difference resulting from nucleotide substitutions (Initial Sequence of the Chimpanzee Genome, Nature 2005)
​

That is their cited source material, the comparison is base pairs, NOT NUMBER OF EVENTS. The number of events does not change the percentage, it's explicitly stated in the paper. No Creationist would get away with such an obvious misstatement, accidental, intentional or otherwise.

The question is simple, did Talk Origins get this statement right, yes or no?

There is nothing complicated about this, it's as simple as 3 plus 1.23, there is no way it's between 1 and 2 percent. Not once have I seen an evolutionist honestly admit this statement is obviously in error. If I can't trust someone with the obvious, why would I take them seriously with the obscure?

Grace and peace,
Mark

 

[mark kennedy]

Are you guys seriously not going to refute me on this because if I'm wrong it's all too easy.

 

[essentialsaltes]

The problem is that there is no unambiguous way to 'do the math'

Compare

abcdeabcdeabcdeabcdeabcdeabcdeabcdeabcdeabcdeabcdeabcdeabcdeabcdeabcdeabcdeabcde

and

abcdexyzabcdeabcdeabcdeabcdeabcdeabcdeabcdeabcdeabcdeabcdeabcdeabcdeabcdeabcdeabcde

If you go position by position, the first 5 are the same and all the rest are different. They are hardly alike at all!

If you step back, and say the second one is exactly the same as the first, except that "xyz" has been inserted at one locus, then how many differences are there? 1? 3?

As we get more complicated with additional insertions and deletions, the two genomes will be of different lengths. How do you even calculate a percentage, when there are two different denominators?

As TO points out, there are still further measures that could be used. Do you compare only functional segments? Do you compare only finished proteins?

So, there are many different ways to measure the difference. None is obviously correct.

Now, as we turn to what TO says, they do not assert what the difference is.

So they cannot have gotten it wrong.

TO addresses the claim: "For years, evolutionists have hailed the chimpanzee as "our closest living relative" and have pointed out that the DNA is 98 to 99 percent identical between the two. Scientists now say the difference is 4 percent, double what they have been claiming for years."

TO does not claim that it is true that the DNA is 98 to 99 percent identical.
TO points out, as I have, that there is no one way to measure the difference.
The fact that two different measures give two different answers should not surprise us.
Therefore the creationist claim, as presented, is deceptive.

 

[mark kennedy]

The problem is that there is no unambiguous way to 'do the math'

Compare

abcdeabcdeabcdeabcdeabcdeabcdeabcdeabcdeabcdeabcdeabcdeabcdeabcdeabcdeabcdeabcde

and

abcdexyzabcdeabcdeabcdeabcdeabcdeabcdeabcdeabcdeabcdeabcdeabcdeabcdeabcdeabcdeabcde

If you go position by position, the first 5 are the same and all the rest are different. They are hardly alike at all!

If you step back, and say the second one is exactly the same as the first, except that "xyz" has been inserted at one locus, then how many differences are there? 1? 3?

As we get more complicated with additional insertions and deletions, the two genomes will be of different lengths. How do you even calculate a percentage, when there are two different denominators?

As TO points out, there are still further measures that could be used. Do you compare only functional segments? Do you compare only finished proteins?

So, there are many different ways to measure the difference. None is obviously correct.

Now, as we turn to what TO says, they do not assert what the difference is.

So they cannot have gotten it wrong.

TO addresses the claim: "For years, evolutionists have hailed the chimpanzee as "our closest living relative" and have pointed out that the DNA is 98 to 99 percent identical between the two. Scientists now say the difference is 4 percent, double what they have been claiming for years."

TO does not claim that it is true that the DNA is 98 to 99 percent identical.
TO points out, as I have, that there is no one way to measure the difference.
The fact that two different measures give two different answers should not surprise us.
Therefore the creationist claim, as presented, is deceptive.

The problem is that the people who did this, many of the same people who did the human genome project. They say 3% and 1.23%, this isn't all that complicated. I'm not asking for an explanation, I've been reading this stuff for years and I know how they do a genomic comparison. I'm talking about a direct statement, a straight forward question requiring a simple yes or no. Did Talk Origins get the math right, because if they did, the Chimpanzee Genome Consortium is wrong.

Two different answers are not an option because this statistic is perfectly consistent with the present research and I could cite and quote half a dozen publications saying exactly what the Chimpanzee Genome Consortium said.

 

[sfs]

Speaking as the local representative of the Chimpanzee Genome Consortium, I'd say they're both wrong. I agree, it doesn't make sense to calculate the number of indel events as a fraction of the genome, since the genome isn't made of events.

On the other hand, I also don't think it makes sense to add the unique sequence in the human genome to the unique sequence in the chimpanzee genome and call that number a "difference". As I've pointed out previously, if 60% of each genome were unique and 40% identical, by this measure the genomes would be 120% different, a number that makes no sense.

If I had to pick a single number, it would 2.7%. That's the fraction of the human genome that is different from the chimpanzee genome (and vice versa). It's also not one of the choices in your poll.

But I've told you all of this more than once before.

 

[pshun2404]

The problem is that there is no unambiguous way to 'do the math'

Compare

abcdeabcdeabcdeabcdeabcdeabcdeabcdeabcdeabcdeabcdeabcdeabcdeabcdeabcdeabcdeabcde

and

abcdexyzabcdeabcdeabcdeabcdeabcdeabcdeabcdeabcdeabcdeabcdeabcdeabcdeabcdeabcdeabcde

If you go position by position, the first 5 are the same and all the rest are different. They are hardly alike at all!

If you step back, and say the second one is exactly the same as the first, except that "xyz" has been inserted at one locus, then how many differences are there? 1? 3?

As we get more complicated with additional insertions and deletions, the two genomes will be of different lengths. How do you even calculate a percentage, when there are two different denominators?

As TO points out, there are still further measures that could be used. Do you compare only functional segments? Do you compare only finished proteins?

So, there are many different ways to measure the difference. None is obviously correct.

Now, as we turn to what TO says, they do not assert what the difference is.

So they cannot have gotten it wrong.

TO addresses the claim: "For years, evolutionists have hailed the chimpanzee as "our closest living relative" and have pointed out that the DNA is 98 to 99 percent identical between the two. Scientists now say the difference is 4 percent, double what they have been claiming for years."

TO does not claim that it is true that the DNA is 98 to 99 percent identical.
TO points out, as I have, that there is no one way to measure the difference.
The fact that two different measures give two different answers should not surprise us.
Therefore the creationist claim, as presented, is deceptive.

To actually confirm a "deletion" one would have to show in the same organisms genome, a time when something was there and now is not (hence being deleted). The same is true for most "insertion" events. The onus probandi requires there be an example of when something that was not there now is (for example, SOME items called ERVs demonstrate this without doubt yet many do not and are justifiably questioned by some even though most simply blindly accept them as insertions).

 

[pshun2404]

Speaking as the local representative of the Chimpanzee Genome Consortium, I'd say they're both wrong. I agree, it doesn't make sense to calculate the number of indel events as a fraction of the genome, since the genome isn't made of events.

On the other hand, I also don't think it makes sense to add the unique sequence in the human genome to the unique sequence in the chimpanzee genome and call that number a "difference". As I've pointed out previously, if 60% of each genome were unique and 40% identical, by this measure the genomes would be 120% different, a number that makes no sense.

If I had to pick a single number, it would 2.7%. That's the fraction of the human genome that is different from the chimpanzee genome (and vice versa). It's also not one of the choices in your poll.

But I've told you all of this more than once before.

And I see it as more like 5% (or more if you consider the additional few percent in the non-coding portions of the genome) and that was not a choice either...

 

[sfs]

To actually confirm a "deletion" one would have to show in the same organisms genome, a time when something was there and now is not (hence being deleted). The same is true for most "insertion" events. The onus probandi requires there be an example of when something that was not there now is (for example, SOME items called ERVs demonstrate this without doubt yet many do not and are justifiably questioned by some even though most simply blindly accept them as insertions).

I'd say that if someone wants to argue that they're not insertions and deletions, they would have the burden of showing that there is some other explanation for why they look like insertions and deletions. As I recall, you faced the same challenge with single-base substitutions and failed to come up with one.

 

[mark kennedy]

Speaking as the local representative of the Chimpanzee Genome Consortium, I'd say they're both wrong. I agree, it doesn't make sense to calculate the number of indel events as a fraction of the genome, since the genome isn't made of events.

On the other hand, I also don't think it makes sense to add the unique sequence in the human genome to the unique sequence in the chimpanzee genome and call that number a "difference". As I've pointed out previously, if 60% of each genome were unique and 40% identical, by this measure the genomes would be 120% different, a number that makes no sense.

If I had to pick a single number, it would 2.7%. That's the fraction of the human genome that is different from the chimpanzee genome (and vice versa). It's also not one of the choices in your poll.

But I've told you all of this more than once before.

Well that sounds about right Steve, but the question is did Talk Origins get their math right. It's really no big deal, but that was the question and the paper was pretty straight forward about it.

This difference corresponds to ~3% of both genomes and dwarfs the 1.23% difference resulting from nucleotide substitutions (Initial Sequence of the Chimpanzee Genome, Nature 2005)
​

That's what you guys said and the math I get from this is 4.13%.

 

[mark kennedy]

To actually confirm a "deletion" one would have to show in the same organisms genome, a time when something was there and now is not (hence being deleted). The same is true for most "insertion" events. The onus probandi requires there be an example of when something that was not there now is (for example, SOME items called ERVs demonstrate this without doubt yet many do not and are justifiably questioned by some even though most simply blindly accept them as insertions).

What you actually have here is a lot like two strings of beads. When you have something in one and not the other it's either an insertion in the one that has them or a deletion in the one that doesn't. Of course that assumes common ancestry, but that's the basic idea. Mind you, some of these 'indels', actually better described as gaps, can be over a million base pairs long.

Human-chimpanzee divergence in 1-Mb segments across the genome.

 

[pshun2404]

I'd say that if someone wants to argue that they're not insertions and deletions, they would have the burden of showing that there is some other explanation for why they look like insertions and deletions. As I recall, you faced the same challenge with single-base substitutions and failed to come up with one.

I do not recall being faced with single-based substitutions as ever being a challenge I was asked to give an explanation for but why do we call them "substitutions" in the first place? To be confirmed to actually be substituted one would have to show they once were one thing, and now are another, and I have not seen this demonstrated in either the human or chimp genome (just that one differs froom the other in these places).

And no the burden is not on someone to show they are NOT (that is asking someone to prove a negative which is illogical). If the claim is made that they ARE this then the burden is theirs to demonstrate it (not just assume it based on a presupposed hypothesis or belief).

I see them as "differences" because they are "different" in each respective genome. Now just because a story is attached to that actual data that explains them in terms of an undemomnstrated common ancestor that's all well and good, as long as we recognize it is only an explanation and therefore no better than any other undemonstrated explanation.

Everybody should watch this video (and this author does accept the idea of a common ancestor but questions the idea of similarity and difference as we have popularized it).

 

[sfs]

Well that sounds about right Steve, but the question is did Talk Origins get their math right.

What part of "I'd say they're both wrong" was unclear?

 

[mark kennedy]

What part of "I'd say they're both wrong" was unclear?

Now that's classic, the larger point was that Talk Origins was equivocating events with the comparison of base pairs. But ok Steve, I think your more then qualified to have a strong opinion that disagrees with both Talk Origins and the paper you helped produce. No problems here.

 

[mark kennedy]

I do not recall being faced with single-based substitutions as ever being a challenge I was asked to give an explanation for but why do we call them "substitutions" in the first place? To be confirmed to actually be substituted one would have to show they once were one thing, and now are another, and I have not seen this demonstrated in either the human or chimp genome (just that one differs froom the other in these places).

And no the burden is not on someone to show they are NOT (that is asking someone to prove a negative which is illogical). If the claim is made that they ARE this then the burden is theirs to demonstrate it (not just assume it based on a presupposed hypothesis or belief).

I see them as "differences" because they are "different" in each respective genome. Now just because a story is attached to that actual data that explains them in terms of an undemomnstrated common ancestor that's all well and good, as long as we recognize it is only an explanation and therefore no better than any other undemonstrated explanation.

Everybody should watch this video (and this author does accept the idea of a common ancestor but questions the idea of similarity and difference as we have popularized it).

Well I enjoyed the video, I think there is something important to consider here. DNA comes in base pairs, there are four nucleotides and two matched nucleotides which are A-T or C-G. Yea, we can be fifty percent bananas but at some point you have to start at functional things like protein coding genes and regulatory sequences.

 

[essentialsaltes]

Talk Origins does not provide a number, so what math did they either get right or wrong?

 

[pshun2404]

What you actually have here is a lot like two strings of beads. When you have something in one and not the other it's either an insertion in the one that has them or a deletion in the one that doesn't. Of course that assumes common ancestry, but that's the basic idea. Mind you, some of these 'indels', actually better described as gaps, can be over a million base pairs long.

Human-chimpanzee divergence in 1-Mb segments across the genome.

"When you have something in one and not the other it's either an insertion in the one that has them or a deletion in the one that doesn't."

Except if they ever were always a normal part of each respective genome just as they are, one in one order with their sequence and the other in their order with a different sequence. If it was always what it is in each (making each creature what they are uniquely), then their is not need to automatically assume one is inserted and the other deleted.

One has to ASSUME that at one time their genomes were the same (even if that is a Common Ancestor) to see substitutions as mutations from and original model. Now to confirm it one must produce an example of the original model for comparison or else the whole assumption remains in the realm of the theoretical and hypothetical. You CAN see that makes sense can't you? See my next post for examples in other fields and maybe that will help....

 

[pshun2404]

In ALL cases of contract law any claims of insertions or deletions must be demonstrated to assure confirmation. The person or persons making the claim of an insertion or deletion, to demonstrate credibility of the claim, must produce an example of the earlier copy for comparison.

If it can be seen that something previously not there now is, or that something previously there now is missing, then it is considered an insertion (to have been inserted) or a deletion (to have been deleted).

How can one prove such a deletion in a data set? If we have access to the complete database by comparing earlier versions with the latest version we can easily detect a deletion with assurance. One can never assume a deletion or call it a deletion without comparison to the earlier version. Hearsay, and opinion, regardless of alleged expertise, is not confirmation of the claim.

In accounting systems the same rule applies. If one claims an insertion or deletion into the record (the cooked books problem) has taken place, confirmation can only be demonstrated when the auditor or examiner is allowed to see or discovers the earlier version which does or does not contain the insertion or deletion, and compare.

MacDonalds was once questioned on their claim that all their food was natural. After investigation it could be shown against the genome of a true actual russet potato that an insertion had occurred. They had inserted a segment of butterfly gene into their potato crops to fight a particular pestilence. But the insertion had been confirmed and so now they are quite honest about it.

In choosing a cloning vector, they must be small molecules because they are easier to manipulate. The sequence being inserted must be capable of prolific replication inside the recipient cell in order to enable the amplification of the inserted donor fragment. Identifying such an insertion is easily confirmed. It is confirmed because we can see how it was not there, and now is there! The same is true when we splice out a segment from an extant genome. We can confirm this actually happened by comparing the genome which previously contained the segment, with the same genome from which it is now deleted. Otherwise such a claim may be interpreted to suggest it happened, and dozens could even claim it happened, but that does not confirm it.

So the burden is no on me to show it did not happen as the historical narrative attached claims (prove a negative) it is on them to prove this is what actually happened.

 

[mark kennedy]

Talk Origins does not provide a number, so what math did they either get right or wrong?

Insertion and deletion (indel) events account for another approximately 3 percent difference between chimp and human sequences, but each indel typically involves multiple nucleotides. The number of genetic changes from indels is a fraction of the number of single-nucleotide substitutions (roughly 5 million compared with roughly 35 million). So describing humans and chimpanzees as 98 to 99 percent identical is entirely appropriate (Chimpanzee Sequencing 2005). (Talk Origins, Claim CB144)
​

They did provide a number, they provided the number of events. It has nothing to do with the total number of base pairs involved but they changed the DNA that is identical to 98% to 99% which is absurd.

Ok here are the numbers from the paper they cite:

On the basis of this analysis, we estimate that the human and chimpanzee genomes each contain 40–45 Mb of species-specific euchromatic sequence, and the indel differences between the genomes thus total ~90 Mb. This difference corresponds to ~3% of both genomes and dwarfs the 1.23% difference resulting from nucleotide substitutions; this confirms and extends several recent studies63, 64, 65, 66, 67. Of course, the number of indel events is far fewer than the number of substitution events (~5 million compared with ~35 million, respectively).
​

Do note that the paper clearly states that, 'this confirms and extends several recent studies', citing five other papers saying pretty much the same thing.

 

[pshun2404]

Another "not accounted for" difference which ups the % are the presence of the genes SRGAP2 (a, b, c, and d) which are located on three distinct parts of Chromo 1 (which in humans are critical to brain development) are nowhere present in ANY apes, chimps included! IMO this separates us from ANY ape-like ancestor (just my opinion).

 

[mark kennedy]

In ALL cases of contract law any claims of insertions or deletions must be demonstrated to assure confirmation. The person or persons making the claim of an insertion or deletion, to demonstrate credibility of the claim, must produce an example of the earlier copy for comparison.

If it can be seen that something previously not there now is, or that something previously there now is missing, then it is considered an insertion (to have been inserted) or a deletion (to have been deleted).

How can one prove such a deletion in a data set? If we have access to the complete database by comparing earlier versions with the latest version we can easily detect a deletion with assurance. One can never assume a deletion or call it a deletion without comparison to the earlier version. Hearsay, and opinion, regardless of alleged expertise, is not confirmation of the claim.

In accounting systems the same rule applies. If one claims an insertion or deletion into the record (the cooked books problem) has taken place, confirmation can only be demonstrated when the auditor or examiner is allowed to see or discovers the earlier version which does or does not contain the insertion or deletion, and compare.

MacDonalds was once questioned on their claim that all their food was natural. After investigation it could be shown against the genome of a true actual russet potato that an insertion had occurred. They had inserted a segment of butterfly gene into their potato crops to fight a particular pestilence. But the insertion had been confirmed and so now they are quite honest about it.

In choosing a cloning vector, they must be small molecules because they are easier to manipulate. The sequence being inserted must be capable of prolific replication inside the recipient cell in order to enable the amplification of the inserted donor fragment. Identifying such an insertion is easily confirmed. It is confirmed because we can see how it was not there, and now is there! The same is true when we splice out a segment from an extant genome. We can confirm this actually happened by comparing the genome which previously contained the segment, with the same genome from which it is now deleted. Otherwise such a claim may be interpreted to suggest it happened, and dozens could even claim it happened, but that does not confirm it.

So the burden is no on me to show it did not happen as the historical narrative attached claims (prove a negative) it is on them to prove this is what actually happened.

Oh but this does get interesting when you start getting into it, apparently at least one chromosome has suffered a net lost of DNA since the split in both lineages:

Calculations from the indels in the 300–5,000-bp range indicate that both chromosomes have undergone a net loss in size since speciation despite frequent insertion events: HSA21q has gained 32 kb but lost 39 kb, whereas PTR22q has gained 25 kb and lost 53 kb. (DNA sequence and comparative analysis of chimpanzee chromosome 22)
​

The human lineage also managed to pick up sixty brand new (de novo) brain related genes:

The de novo origin of a new protein-coding gene from non-coding DNA is considered to be a very rare occurrence in genomes. Here we identify 60 new protein-coding genes that originated de novo on the human lineage since divergence from the chimpanzee. The functionality of these genes is supported by both transcriptional and proteomic evidence. RNA– seq data indicate that these genes have their highest expression levels in the cerebral cortex and testes, which might suggest that these genes contribute to phenotypic traits that are unique to humans, such as improved cognitive ability. Our results are inconsistent with the traditional view that the de novo origin of new genes is very rare, thus there should be greater appreciation of the importance of the de novo origination of genes…(De Novo Origin of Human Protein-Coding Genes PLoS 2011)
​

Less is more I guess.