Any time the various creationist organizations — AiG, ICR, CMI, DI, etc. — start getting excited and claiming that genetics supports creationism, it usually seems to trace back to Jeffrey Tomkins, the one guy who knows a little genetics and molecular biology, and most importantly, knows how to distort the scientific literature. A new paper in Nature, the complete sequencing of ape genomes, does a detailed and thorough comparison of great ape genomic data, and Tomkins does his usual thing and butchers it.
Tomkins is known for his usage of “ungapped” comparisons to depress the percentage similarity between the human and chimpanzee genomes. This method relies on aligning the beginnings of two DNA sequences, and measuring whether subsequent base pairs at corresponding positions match one another. The flaw in this method is that insertions, duplications or deletions in either sequence may cause parts of it to be shifted forward or backward relative to the other, so that equivalent sets of base pairs are not precisely aligned with one another in the comparison. Ungapped comparisons interpret those parts of the two sequences as entirely mismatched even if there are no other differences between them.
If you see any creationist now claiming that humans and chimpanzees are 15% different, rather than the number reported in scientific journals of 1.5%, it’s all coming from the mangled misinterpretations of Tomkins, who really is obsessed with the idea that humans can’t possibly be at all related to other apes. Casey Luskin accepts the distortion and is stating that scientists have been hiding the magnitude of the differences.
They haven’t. The root of the problem is that there are multiple ways to compare sequences of 3 billion nucleotides. One way is to compare aligned sequences, that is, the genes and regulatory stuff that makes up the functional bits of the genome, and there you find about 98.5% similarity between chimps and humans. Another approach is to tally up all of the sequence differences, whether they have any phenotype or not, and there you can find all kinds of repetitive, noisy stuff in the genome. You can find that a human parent is 10% different from their own child! Here’s a good explanation of the whole data set, rather than a Tompkins-ish cherry-picked mess of lies.
Not mentioned, unfortunately, is the ultimate key to explaining these differences: the differences are in the genetic junk. I guess it’s fair to not bring that up, since creationists do not believe in that anyway.
It does expose the fact that ultimately, all the creationist organizations, including the Intelligent Design wackos at the Discovery Institute, do believe that humans were separately created by a deity/aliens. If that wasn’t their endgame they wouldn’t be paying any attention to Tomkins’ nonsense.
I can’t let this pass. Casey Luskin is particularly egregious in claiming that scientists are lying.
These are all groundbreaking findings — and it’s a shame that Nature would not report the data clearly and would make all of this so hard to find — using jargon that most non-experts won’t understand. Why did they do this? It’s important to realize that publishing scientific papers can be a bit like sausage-making: it’s often messy, and the final form that you read usually represents compromise language that all of the authors, reviewers, and editors were willing to publish — and may not represent precisely how every author of a paper feels. So perhaps some authors of this study would have preferred to state the implications more plainly. But we can still ask, Why didn’t Nature state the results clearly and let the chips fall where they may?
Note that this is a response to Nature publishing the complete and detailed results of a complex genetic comparison — they did state the results clearly, and published all of the data. None of the creationist critics have added any new information, every complaint they’ve made is the product of extracting bits and pieces from the Nature paper. It’s not their fault that the paper doesn’t state the implications more plainly
because the creationist implications are not there.
It annoys the hell out of me that Nature can publish a 28 page paper with 82 tables of data in the supplementary information, and Luskin can whine that they didn’t dumb it down enough that a lying creationist can find the part where real scientists say god did it
.
It’s because the data don’t support your claim, you ass.
I used to find this stuff hilarious. Now I just get bored. These days, I crave flat-Earth creationists who worry about the shape-shifting reptiles while searching pizza places for non-existent dungeons.
BTW are there many non-repetitive non-phenotype genes? I assume a lot are recessive genes and fragments of viral genomes but I don’t know all categories
Even using Tomkins’ flawed method, genomes will show increasing similarity between more closely related species, which is evidence of common descent and not special creation. His claim has nothing to do with analysing evidence, it’s about poisoning the well so his followers won’t accept findings from scientific papers.
Not quite. 98.5% similarity (or 98.77%, as in the original chimp genome paper) comes from looking at all alignable sequences, most of which are not functional. Around 85% of the genomes are alignable. What Casey does is count an indel of, say, 1000 bases as 1000 differences. But at least, unlike Tomkins, he actually uses an alignment.
@birgerjohansson: You need to find Larry Moran’s book What’s in Your Genome, which will explain all the categories of DNA for you. There’s no such thing as a recessive gene, just recessive alleles; you may be thinking of pseudogenes.
@chrislawson: Rather than simple similarity, think nested hierarchy, While the two often go together, they don’t have to, and it’s the latter that shows common descent.
Clarifying the Tomkins method
Original text: To be or not to be, that is the question.
Misprint A
Xo be or not to be, that is the question.
1 substitution error in 40 characters
Scientific standard: 97.5% similarity
Tomkins: 97.5% similarity
Misprint B
o be or not to be, that is the question.
1 deletion error in 40 characters
Scientific standard: 97.5% similarity
Tomkins: 0% similarity
John Harshman@4–
Thx for the clarification.
Related to? We are apes. Just one of many species of them.
I dunno, but it seems to me that addressing tomkins claims, say, as one would a scientists claims and then dismissing them the same way is flawed. Because they are NOT scientific claims. They are made up bullshit. To poke holes in the 15% claim, say, is to give the claim too much credit. After all, according to him, were all different kinds.. So hit him there! Like, if were all different kinds, why do we all have genes at all? As opposed to we’ve got spiral gene stuff, but they’re a different kind so they should have cubic silicon based let’s call them henes, vs the whatever you creationists expect as a difference in that other kind of marine thing cos reasons vs… Why are we all even cellular, right? Despite plants photo-synthesising for a living they’re still (not a biologist, but whatever it is, right?).
To look at the different living things is to look at things SO very different that it took us a long time to classify some of them as living at all. Yet even creationists acknowledge the sameness, the gene-iness, of all living things, but why? Why do they acknowledge that at all? Why, according to them, do kinds have these similarities, yet maintain the profound differences they claim must exist? If you are comparing a forest to a coral reef, as they claim we are cos kinds, 15% seems too tiny a claim for them to make. I would say to them that because of what they claim it should be a massively larger difference, and then therefore ask them to explain why it isn’t.
If the above is confused, I apologize, I am not a biologist. But I hope the point is clear.
Another clarification: the methods Tomkins and Luskin used to come up with their numbers are entirely different, even though they managed to arrive at similar conclusions. They’re nonsensical in different ways, though I’d say that Tomkins’ method is the more ridiculous of the two. @chrislawson in #5 explains it. Luskin’s method, on the other hand, has actually been used on rare occasions by actual scientists.
So, to paraphrase a stupid and meaningless old argument, if humans evolved from monkeys, why are there still ignorant monkeys like this guy?
This method relies on aligning the beginnings of two DNA sequences, and measuring whether subsequent base pairs at corresponding positions match one another. The flaw in this method is that insertions, duplications or deletions in either sequence may cause parts of it to be shifted forward or backward …
Huh. Reminds me of a typing-test program I tried on an Apple //+. Fingering as fast as I could, I missed one letter – but the software scored all of the (correct) entry after that as wrong, because none of it matched the nth-letter characters in the original text. I hope no one ever used that app to rank job applicants (or Tompkins to grade students).
indianajones@8–
Tomkins is making a scientific claim, based on his calculation of genome similarities derived from published data. I believe it is important to point out the error of his calculation method and interpretation as well as the non-scientific, creationist-apologetic aspects of his argument.
@12 chrislawson
Ya might well be right, I’m not an expert. But I don’t argue against angels causing gravity by pulling everything to the ground on the basis of the implausibility of their wing aero-dynamics, even though wing aero -dynamics are a real thing. Not when the angels themselves are stupid and impossible. Credit where it aint due and all
I don’t get it. What kind of creationist is going to accept that we have 85% of our DNA in common with chimpanzees? I use that 1,5% tidbit to shock agnostics into realisation, but for a creationist, everything bigger than zero should surely be unthinkable?
indianajones@13–
Funny you should say that!
Feynman, of course, was not actually advocating a modified angel hypothesis, and I agree with you that Tomkins’ underlying belief system is as ridiculous as the angel theory of orbital mechanics. But Tomkins made a specific claim of fact about genetic similarity, that claim is demonstrably wrong, and I am happy for people to demonstrate it. And while I also would not engage in the minutiae of angel wing aerodynamics as it applies to orbits, I would ask anyone who presented this hypothesis how wings are supposed to apply thrust in the vacuum of space.
taemon@14–
That’s the whole point. Tomkins is not interested in presenting a coherent argument. This is just shouting “They exaggerated! You can’t trust anything scientists say!” to his base. There’s a disturbingly funny scene in Jim Thompson’s novel Pop. 1280 that describes this strategy perfectly.
The whole point of Tomkins’ “work” is to make creationists feel better about believing in their particular fairy tale. “Scientific rigor” is simply not relevant, and credibility is only relevant in terms of it helping his intended audience (creationists) rationalize their decision to pay attention to him. He’s peddling misinformation, not genetics.