According to my googling, geneticist Jeff Tomkins has been mentioned before in this website once, by Cordova in 2015. Nobody cared back then. Now here is a recent interview with him https://www.youtube.com/watch?v=Vxk1dZrnBR8 (it’s audio rather than video)
In this interview, Dr. Tomkins states that the claim of 98% similarity between humans and chimpanzees is only based on certain “regions of DNA”, i.e. some sequences, not the whole genome. Further specific claims:
- The similarity between humans and chimps, taking the whole genome into account, would be rather 70% or two thirds
- For evolution (of humans and chimps from a common ancestor) to be true, 98% similarity (across the whole genome) would be required, given the rate of mutation
To those who know better: Are these two claims true? Certainly evolutionary biologists would not be so sloppy as to declare 98% similarity of something without proper justification!
Further, at 15m20 mark, Dr. Tomkins says that genes operate in “networks and subnetworks,” apparently so that the whole genome is as if an integral system or organism by itself. This would imply that the genome would be able to evolve/change not by mutation and natural selection, but as predetermined by the inherent nature of the genome.
At 22m50 mark, Dr. Tomkins says, “Every time an unusual creature has its genome sequenced, we are finding unique sets of genes for these creatures.” Apart from the evidently unscientific term “creature,” I am interested in specific examples. Dr. Tomkins mentions “orphan genes” of shrimps, oysters and insects in that section.
How do evolutionary biologists assess Dr. Tomkins’ performance in this interview and his credentials in general?
Allan Miller,
DNA_Jock,
Are humans 98% chimp? Or 70%?
The basis of the op was discussing limited types of measurement of differences delivering a biased result. Excluding how differentiated proteins are generated in embryo development would appear to be on subject.
Yes, the comparison is between putatively orthologous loci. The pr nucleotide identity score. If we take some arbitrary stretch of 1000 nucleotides from the human genome, how similar will the most similar stretch from the chimp genome be? Turns out the answer is roughly 99%.
What those 1% differences do is irrelevant to whether it is in fact the case that the pr nucleotide identity score is 99%, or 70%. Those 1% differences could certainly result in different gene expression patterns, different patterns of splicing (whether those are functionally important or not), and so on. That doesn’t change the fact that when we do those DNA sequence-to-DNA sequence comparisons, the identity score is roughly 99% on average.
It should be noted that if it actually WERE 70%, it would not make a difference with respect to our ability to infer common descent. There are in fact species where the average identity score between humans and that species is 70%, and we are related to those too. All it would change is how far back our common ancestry goes, that’s it.
As has been explained a ridiculous number of times before, evidence for common ancestry comes from statistically significantly high amounts of tree-like structure in the similar genetic sequences from different species, and the fact that independent phylogenies overwhelmingly corroborates a similar branching topology to an extent that yields the best agreement between prediction of theory and observation in the history of science.
Isn’t Tomkins the guy who calculated similarity by fragmenting the genome into very small pieces, then counted only pieces that completely matched a piece in the other genome as similar, rejected any other piece? The no partial matches guy?
Certainly. I think many people here would agree that organisms cannot evolve by degradating or breaking gene function, because that would collapse the entire system.
No, it wouldn’t. The percentage similarity is a genomic statistic, and is not affected by what happens during development. We aren’t ‘x% chimp’ any more than they are ‘x% human’. We have diverged from a common ancestor to the extent that we are x% similar, down from c100% at the moment of bifurcation.
Rumraket,
The question is can you get from path a to path b with reproduction and natural variation alone? Large changes in protein sequences (alternative splicing) that is not planned maybe problematic.
Huh? At some point in time, all human and chimp ancestors had the same genome?
Does that mean all chimps have the same genome now?
Why yes, yes he was!
WTF. That’s not what he said
dazz,
What did he mean by 100% then? 70%?
You’ve already decided the answers to all such questions is “no” so why bother with the question mark?
Yes, and then things die. And that “change” is also dead.
What evidence, other then personal incredulity, do you have that such changes were intelligently designed?
Just to reiterate what others have said, the mutation rate and evolutionary change are two separate things.
Mutations (changes in the DNA sequence in the genome) can occur in several ways. It is more accurate to talk about the mutation rate per generation than vs absolute time. So the LTEE experiment has run for around 50,000 generations of E coli in one human generation. Without variation, natural selection could not happen. The rate of evolutionary change depends on whether new variation (alleles – competing DNA sequences for the same position in the genome) is beneficial, neutral or deleterious to the organism with respect to its niche (the space the population occupies).
In a stable niche, deleterious alleles will be selected against (purifying selection) and populations may evolve very slowly. Mutations that occur in non-functional DNA sequences will undergo no selection and will tend to hang around (drift is an additional factor that tends to reduce diversity randomly but that’s a detail) and this background rate of variation can be used as a molecular clock to estimate the time of separation between species.
I would say it’s an approximation or a simplification, just like “the moment of divergence” should not be taken literally. Back when both populations were one, there would still be some variation, of course. Maybe something like a .1% which I believe is what we now observe in humans
At the time of separation of one population of the ancestral species into the population that led to modern humans and that resulting in modern chimps, the gene pool was c100% similar ( c means approximately). When gene flow ceased between the two populations, differences could creep in and now the two genomes are at least 1% different.
No.
ETA “similar”
He seems to describe the similarity calculations in this way in the interview (except that perhaps he calculated the non-matching pieces as no match). Are you saying this is the wrong way to do it? What is the right way?
He explicitly believes this: The phenomenon of biological life did not develop by natural processes from inanimate systems but was specially and supernaturally created by the Creator. so why should anyone care what he thinks about how related one species is to another when they are all totally unrelated anyway.
https://www.icr.org/tenets
Erik, There’s a good analogy that Richard Dawkins uses (I think it’s in The Ancestor’s Tale) comparing molecular phylogenics to textual analysis to establish the hierarchy among different early manuscripts of Chaucer’s Canterbury Tales. I’m sure it’s been mentioned here before. I’ll have a look for a reference.
Erik,
He is wrong on a number of levels. He has something of a history in this regard.
just glancing through a PDF of The Ancestor’s Tale, I can’t seem to find it. Bear with me.
ETA starts at p111
Ah, Glenn Williamson has the Australian accent, presumably. What an impressive takedown. Has Tomkins ever responded to having the software bug pointed out to him?
I see Tomkins has published a later paper (Comparison of 18,000 De Novo Assembled Chimpanzee Contigs to the Human Genome Yields Average BLASTN Alignment Identities of 84%) at Answers in Genesis now claiming 84% similarity.
Is this his final answer?
Alan Fox,
Quoting the final paragraph:
A glaring 20% overall DNA similarity difference between the human and chimpanzee genome is an evolutionary discrepancy that cannot be dismissed. This extreme level of genetic discontinuity raises serious issues for the evolutionary myth that humans and chimpanzees share a common ancestor not more than about 3 to 6 million years ago which largely depends on a 98 to 99% DNA similarity to seem theoretically possible. The uniqueness of mankind as stated in Genesis, “So God created man in His own image; in the image of God He created him; male and female He created them,” (Genesis 1:27 (NKJV)) is now soundly confirmed by the scientific data.
In 2010, a poster at Uncommon Descent called “niwrad” made that mistake, coming up with a 35% difference between humans and chimps. I responded at Panda’s Thumb here, showing that there is no conflict between the two figures.
It depends on how you count gaps of different sizes in either sequence.
Glenn Williamson’s response to Tomkins was on that same issue and fairly well demolished Tomkins’ figure.
Alan Fox,
The recent “18,000 fragments” paper is what caused the video that I linked to originally. Evidently there’s quite the history here – going back to the BLAST bug and the utterly nonsensical results that that generated.
I do remember the ‘exact matches’ fiasco, but I can’t find a reference to it…
ETA: Aha!. Thank you Joe.
ROFL. How is the degree of similarity in nucleotide sequence between human and chimp in any way an indication about whether mankind was made in God’s image?
🙂
Unlike a partial non sequitur.
Evolution by devolving. Isn’t that the YEC model though? Bayesian analysis indicates that lots of people probably believe it.
The evolutionist argument seems to be (once again):
-Tomkins numbers are wrong.
-The numbers don’t really matter.
-Even if they are right so what.
-Its not a problem for evolution.
-We knew this all along.
-Exactly as we predicted.
I think he meant 98% and just rounded up a bit.
Its interesting that Tomkins uses the mutation rate to suggest that the sequence similarity between chimps and humans is a problem. Its not and that’s addressed elsewhere. Its just that in the context of another discussion, such as the existence of Adam and Eve, he’d say mutation rate was very different.
Its like the YEC who challenges the foundational evidence for why radiometric dating works and is a reliable way to determine the ages of geological formations but uses radiometric dating to ‘prove’ the Shroud of Turin is 2000 years old
Mung,
Yea, you are probably right. The differences between chimps is probably pretty similar to the difference between humans and chimps.
Anyway, the numbers don’t matter. They are wrong. Exactly as expected. So what.
Exactly. Not a problem. Just what we expected, 98%..er 70% er, …numbers don’t matter.
Not a problem.
We have gone from VJ claiming the numbers are wrong about predicting events from the Bible, to evolutionists claiming numbers for DNA sequences don’t matter.
Welcome to crazy world.
Tomkins numbers are wrong and that does matter
The numbers only dont matter when discussing the comparative relatedness between humans and a variety of other organisms
The numbers aren’t right.
Its not a problem for determining which organisms are more closely related to humans. It would be a problem relative to the mutation rate AND the estimated divergence from Chimps based on the fossil record.
No one says they new this all along
No one attempted to predict how similar chimps would be to humans at the DNA level. At the time there was disagreement over whether chimps or gorillas were more closely related humans. So most people would have predicted those 2 being the 2 closest in DNA sequence but no one would have attempted numbers or %
That’s a misrepresentation. Vincent just thinks it’s possible to make a calculation.
That’s a misrepresentation, too. The method adopted in calculating similarity can significantly alter the result.
Apparently not to Tomkins. 70% has changed to 84%
Ironic that even the creationists are able to put their reputations, such as they are, on the line and say something and here you are saying nothing, as usual.
That’s not the question.
No. The ‘c’ preceding ‘c100%’ is short for ‘circa’, a common means of indicating approximation.
Nope. I meant what I said. c100%. The c is significant.
The ‘gotcha’ merchants are busy today.
So would I be right in thinking that the Creationist argument is
– Tomkins disagrees with evolutionists
– Tomkins must be right
?
OMagain,
The amount of splicing changes and number of lethal splicing changes in human cancers.
Allan Miller,
What is the question?
It is literally the title of the OP, albeit rather poorly expressed. What is a sensible figure for the approximate percent difference between those lovable chimps and some random self-absorbed bipeds?
Allan Miller,
Fair enough.
Tomkins has had several bites at this apple, and he has purposefully misled his audience in every case.
The “certain regions” of the genomes that are being compared are the regions that have been assembled. Tomkins falsely asserts that unassembled sequence reads means they don’t have homologous DNA in the other genome. This is false. You can’t compare unassembled regions since you don’t know where they fit into the larger genome. That’s why they are excluded from comparisons, not because they don’t have a match in the other genome.
It is completely wrong to claim that if 10% of the chimp genome is unassembled it indicates that this region has 0% similarity to the human genome. We could also go through Tomkins other complete misrepresentations, such as using an ungapped comparison to get lower numbers.
The sensible way to describe it is:
1. In the DNA regions that humans and chimps share through common descent, about 98% of the bases are still the same. The 2% difference is due to substitution mutations.
2. In total, 96% of the chimp and human genomes are shared through common descent with 2% of that difference caused by insertions or deletions of DNA in either lineage since common ancestry.
Yes, but YECs do not usually grant that breaking stuff in a polyconstrained network increases fitness.