Tomkins’ Chimps

A diagram of the Ape and Human genomes, exaggerated and simplified for effect: see below

The final article from the June Acts & Facts (that I intend to cover individually) is Jeffrey Tomkins’ Journal Reports Bias in Human-Chimp Studies. This is the latest in Tomkins’ crusade to prove that we aren’t really almost identical to apes.

The article is really just an advertisement for two papers authored by Tomkins along with Jerry Bergman in the April Journal of Creation on the subject of Human-Chimp genetic similarity (hence the title). The JoC is not open access, at least not when it is initially published, and so these papers aren’t available. On Tomkins’ blog, Designed DNA, there is a form for asking for these papers – I have asked, but have not received. What is available, however, is Bergman’s “Anders Breivik—Social Darwinism leads to mass murder“: other pieces of nonsense in this edition include John Baumgardner’s “Is plate tectonics occurring today?”, Humphreys’ “The moon’s former magnetic field—still a huge problem for evolutionists”, the rather out of date Jonathan Sarfati’s “Neutrinos faster than light?—will relativity need revising?”, and another article by Sarfati called “Should creationists accept quantum mechanics?” (he says “yes,” if you’re wondering). Micheal Oard also claims that “Post-Flood man is becoming smarter and more human,” something that could prove useful in countering the ‘genetic entropy’ lot.

Being thus unable to get my hands on the actual papers – Is the Human Genome Nearly Identical to Chimpanzee? A Reassessment of the Literature and Genomic monkey business—estimates of nearly identical human-chimp DNA similarity re-evaluated using omitted data if you want titles – I’ll take a moment to tackle Tomkins’ research on this topic in general (though bear in mind that I’m no geneticist).

Tomkins is after the holy grail: to prove that humans aren’t all that related to Chimpanzees, genetically speaking. Personally, I think he’s barking up the wrong tree. Rather than playing with the absolute numbers he should be looking to see if we are relatively more genetically related to Chimpanzees than we are to other animals and whether this similarity was more or less than between two organisms of the same creationist ‘kind.’ That is what counts, when it comes to The Controversy, not the absolute number. But if Tomkins wants to waste his time on it I can’t stop him, can I?

He reported in the January Acts & Facts that he had in fact come up with such a number: “between 86 and 89 percent.” This isn’t really low enough for his purposes, and so he’s now trying to lower it even further:

The first of the recent JOC papers reviews secular science literature associated with the common claim that chimpanzees and humans are nearly identical. This analysis took the published secular claims at face value and showed that many differences exist in regard to not only genomic DNA, but also to gene regulation, regulatory DNA features, microRNA code, and gene splicing. Multiple types of DNA sequence and genetic mechanisms reported in the standard scientific literature clearly show that major genetic differences exist between humans and chimps—features clearly predicted by the creation model outlined in the book of Genesis.

Here, he tries to argue that there are more elements that are different than just genes. The problem here is that he appears to be trying to count things twice: the difference in genetic material will be one of the causes of these other differences. And this whole “clearly predicted by the creation model outlined in the book of Genesis” is bullshit – completely made up. The only genetics in Genesis is in the story of Laban’s cattle, and in that story Jacob actually has no idea what he’s doing (scientifically speaking) and is operating on the principles of sympathetic magic.

The second paper examines the research methods and discarded data reported in an assortment of key secular human-chimp DNA research publications. All analyzed cases of reported high human-chimp DNA sequence similarity are based on biased data selection and exclusion techniques. DNA sequence data that are too dissimilar to be conveniently aligned are omitted, masked, or completely excluded. Furthermore, gap data within DNA sequence alignments are typically omitted, further biasing similarity estimates.

Now, here’s where we get to the diagram I put at the top. In a perfect world, we would be able to reorganise the Human and Chimpanzee genomes like that diagram, with the sections of the genomes that are completely different at the ends. Now, for practical purposes (such as the relative genetic relatedness of different animals) it only really makes sense to compare the stuff between the dashed lines for similarity. I mean, if a mouse had a whole bunch of gunk at the end of its genome enough to make it several times longer than any other that wouldn’t make it completely unrelated to anything else, would it? Or, at least, take the entirety of one genome and see how much of it you can find in the other. If you go further and try to factor in the section of the other genome that doesn’t match up it doesn’t seem to me to make an awful lot of sense to express your outcome as a percentage figure. A percentage of what? You also run the risk, once again, of counting your differences twice.

These highly selective data-discarding techniques, fueled by Darwinian dogma, lead to the commonly claimed 98 percent similarity in DNA between human and chimp. Based on the reanalysis of DNA similarity estimates using discarded data in leading secular research publications, it is safe to conclude that genome-wide DNA similarity between human and chimpanzee is not more than 81 to 87 percent identical. These numbers are in good agreement with the range of estimates obtained by ICR’s independent research.

The problem is that this apparently common-sense system looks to a paranoid creationist like, well, “highly selective data-discarding techniques.” It’s also important to note that, back in January, Tomkins called the “86 to 89 percent” to be “a very conservative estimate of human-chimp DNA similarity” (though he did give himself some wiggle room). He here concludes:

One must keep in mind that the chimpanzee genome is larger than the human genome by at least 8 percent (based on current data). Also, many non-similar regions of DNA between the human and chimpanzee genomes cannot be accurately compared due to high levels of dissimilarity. Therefore, overall genome similarity between human and chimpanzee is most likely lower than 81 percent.

This, however, is clearly ‘counting twice’ (if not more times). His original experiment, after all, consisted of taking fragments of chimp DNA and seeing if he could match them to the human genome. This already corrects for the larger size of the chimpanzee genome. And I’m sure that, if we got a proper geneticist onto this, yet more problems with Tomkins’ research will just pop out of the woodwork.

22 thoughts on “Tomkins’ Chimps

  1. The human accelerated region (HAR) is the part of the human genome which is highly conserved in other animals (i.e. is very similar between say a chimp and a chicken) but is rather different in humans. This would suggest that it is a rather large part of what makes us “human.” Despite being so important they actually make up a relatively small segment of the genome.

    In other words, what the first creationist is crowing about is nothing new. Scientists already understand that despite humans and chimps being very similar genetically those few changes can have rather large effects. Of course, that doesn’t stop them being small changes either. As such they aren’t exactly an insurmountable problem for evolution.

    So basically that first article is redundant, since scientists already understand that theses changes can have large effects and irrelevant since those large effects are still the product of a few changes and so they do nothing to refute the relationship between humans and chimps.

    • But do creationists understand it? I have a feeling that’s a ‘no,’ from past experience.

      Anyway, Tomkins has now offered copies of the paper, so I should get a chance to take a closer look into what he actually wrote, rather than his reporting here.

    • Ha, the answer to “do creationists understand it” is typically no. Anyhoo, it should be interesting to see what an actual creationist “journal” looks like. I look forward to any posts on the subject.

  2. I had a look at the article – “The moon’s former magnetic field—still a huge problem for evolutionists” and I must admit I don’t really get his point. So apparently some rocks gathered from the moon in the ’70s formed in a strong magnetic field, but there isn’t one on the moon. Doesn’t that suggest that they didn’t come from the moon originally rather than that the moon’s magnetic field has changed?

  3. I am missing your point as to why you cannot express the comparison as a percentage figure? If there are, say 1 billion base pairs in the chimp DNA and 1.1 billion in humans then why is it not possible to take the similarities in the two sequences, say 0.9 billion and express it as a percentage of either the chimp or the human – 82% compared to the human or 90% comapared to the chimp?

  4. Doesn’t matter. You can express both ways if you want. Either way we get an understanding of how similar or dissimilar they are.

    • There are two ways that genomes can differ: changes in size (i.e. whole sections on one that just aren’t found in another) and changes in the sequence of the parts found in both. I think the typical practise is to only look at the changes in sequence, as to measure both at once confuses the two. What Tomkins seems to be doing then is going the other way and producing a lower number, but not making clear the extent to which he isn’t actually doing anything new.

      How would you measure the similarity of these two sequences as a percentage?




    • It seems to me that these two sequences are very dissimilar. Eventhough there are only two differences in the first 10 bases there are added bases in the second making the two very dissimilar making the percentage 8/21 = 38%

    • The original sequence I started with was ATCGGAATGA.

      To make the first given I just added a T at the end and mutated one of the bases to produce ATCGAAATGAT.

      To make the second I duplicated it, with a C between them, and mutated a couple more bases to produce: AACGGAATGACATCGGATTGA.

      That’s not a lot of changes, but it produces a similarity by that method that’s not a lot better than two random sequences of the same length. You could do it that way, but I’m not sure the result would be very useful.

      Edit: Changed bolding to underlines, because it wasn’t showing up very well.

    • I understand but if one species have extra bases compared to the other how could you say there is 98% similarity between the two without taking the extra bases into account?

    • I would say that you should probably acknowledge that there are sequences that don’t align, but that you shouldn’t directly include them in the percentage as it gives a less useful result. I’m not exactly sure what the typical scientific approach is though.

      I’ll add that I really don’t like it when people argue for evolution based purely on a raw percentage comparison: a much more nuanced argument can be made on this subject, and should be.

  5. I just read that the total chimpanzee genome has been sequenced and the percentage difference from human is 5% after comparing 779 kilobases. There were 1.4% substitutions and 3.4% additions or deletions (indels). So apparently they do add the substitutions and the indels.

    Another question I have is how are these random changes made to the genome without disrupting the whole sequence let alone to add new features. If the code in any computer program is randomly changed would we expect to get a better program or a disruption of the existing one?

    • I would say that the analogy of a computer code has been oversold. If DNA is a computer program, than this is a program that you can hit with a hammer, to mix metaphors, and be most likely to do nothing at all. Indeed, we know from experiment (e.g. Lenski, but there are others) that it’s perfectly possible for mutations to have beneficial effects. This really isn’t what you would expect from a computer code – even one with massive redundancy, or some other mechanism to protect from damage – and this should tell us that the analogy should not be used to make predictions about how mutations should behave. In short, the map is not the territory.

  6. But the beneficial effects that are seen – if I understand it – are being able to thrive on certain substrates where the wild type would die out and those not resistant to the substrate would live or where some binding capability of a toxin was negated by a mutation. Where have we seen a beneficial mutation that causes a new structure to form or even an organelle in bacteria? I understand that bacteria has been reproduced for almost a million generations (every 20 min for 35 years) under different conditions but with no hint of a new organelle forming or the modification of an existing organelle.

    How does introducing random errors into the DNA code make structures like the neck of the giraffe or the tail of diplodocus when everything we see around us seem to go the other way – order to disorder?

    • Part of that question revolves around what you consider a “new organ.” The neck of a giraffe, after all, is merely a neck – if we invented a time machine and spent many thousands/millions of years in the field, closely watching an Okapi-like creature lengthen its neck until it became a giraffe, I don’t think that would persuade anyone who thought evolution could not produce novel organs. But this is a terrible place to draw the line: the evolution of humans from apes, for example, didn’t require much in the way of new organs, but there are few creationists who are prepared to accept this but not evolution as a whole.

      The Lenski experiment has only been going on for 50,000 generations, so should we expect to find such things in that time? And what would they be? – bacteria aren’t known for having organelles at all. Last year there was a paper about bacteria that, in a matter of weeks, had evolved new flagella, although they were of course duplications of their existing one, working in tandem with each other. Carl Zimmer wrote a nice piece on creationist claims that this apparently wasn’t evolution.

      On the “order to disorder” – do they? Thermodynamics does say that entropy – not the same thing as disorder, though it sometimes can be used as an approximation – must increase in a closed system, but the system that we inhabit is of the universe as a whole. Individual portions of that system are not obliged to follow that progression at all times, so long as the whole does. As a tree dies it’s entropy increases, but when it was first growing it decreased. And there will be many more trees to follow.

  7. Bacteria does have some organelles such as ribosomes and cell walls. But regardless it seems to me that new organelle formation would be expected as eukaryotes have them. I thought i read that over 500,000 generations have been produced (it can be done in 35 yrs) but perhaps I saw an additional 0 from the 50,000. May be it is not enough time so we’ll wait and see what may be produced. It is impressive that new flagella formed – even though the instructions for making them already existed – but do we know that they have never had multiple flagella in the natural environment and that this is not merely normal variation in this bacteria.

    But look at the immense variety of organs and structures and systems there are in iiving things. Are we to really believe that all this came about via mistakes in the reproductive process. OK, entropy must increase in a closed system and entropy could decrease in an open system with input of energy but must not this energy be channeled by an existing mechanism to do so like the photoynthetic apparatus in plants? Do we know of any system that makes entropy to go the other way without an existing mechanism in place?

    • The Lenksi experiment has only about 25 years under its belt and the time for reproduction is a fair bit longer (and more complicated than) a simple 20 mins per generation, which might be where you’re getting the confusion from.

      It does seem, considering how quickly (and apparently repeatedly) they appeared, that the multiple-flagella state is caused by mutations that can and do occur in the wild, but which are not usually beneficial so don’t become common. But in the new environment of the lab circumstances are different and so when they appeared this time they quickly outcompeted the competition. Given more time this duplication could be built on, with the new flagella diversifying in function. If one of them gained the ability to ensnare other bacteria as prey, for example, would you consider it a new organ?

      The errors in reproduction provide the substrate on which other processes, such as natural selection, can operate and mould the organism to its environment. The finished product is analogous to a polished river stone: uncountable random, individually destructive collisions all add up to produce perfection. Looked at from the molecular level it may be hard to believe that this could be so, but the result is plain. On this note: have you ever had a play with an evolution simulator?

      The local decrease of entropy does not require an “existing mechanism” at all, but is a natural and necessary product of thermodynamics. Imagine two closed systems with different levels of entropy – two objects with different temperatures will do – and consider what happens when you bring them together to make one big system. The overall entropy must increase, as we know, and this is also what would be observed from inside the system that started off with less. But for the one that started with more entropy the contact will actually decrease, no mechanism required.

      Earth lies between the powerhouse of the Sun and the cold vacuum of space. This kind of energy flow is naturally going to power something interesting, though on an atmosphere-less rock like Mercury this is harder to see. On Venus and Jupiter, for example, it produces amazing weather patterns; on Mars we see great sand dunes marching across the landscape; on Titan there is an entire water cycle, complete with rivers and lakes, but with organic chemicals instead of water; and on Earth it powers life.

  8. Just to let you know I am not against evolution. I am however skeptical about the whole molecules to man scenario because I think there is no direct evidence for it although there may be circumstantial evidence. I don’t think the evidence presented would have held up in any other branch of science as many scientists have admitted.

    You asked whether I would consider it a new organ formed if the multiple flagella bacteria captured the single flagellum bacteria. If this happened and the captured bacteria formed a new function within the multiple flagella bacterium and if it continued to reproduce like that then I would say a new organelle has formed.

    I am sorry but I don’t think your polished river stone analogy has any connection to forming new organisms. There is no perfection in a polished river stone. We just happen to like the polished product. There is no entropy decrease in the polished product. No, I haven’t played with an evolution simulator but I think these computer models have no bearing on what happens in real life.

    In bringing two closed systems with different levels of entropy you gave you say that the system with greater entropy will end up with less entropy. This is however, very temporary and the one that gained entropy will soon reach the level of the other and then continue to increase in entropy. Also, if you are trying to use this as an analogy in evolution I think you got a problem because the environment has greater entropy than the cell and thus the cell would have to give up its lower entropy to the environment. So the environment would decrease in entropy and the cell would increase!

    On the sun giving energy to the planets: Here again I don’t think there is any connection to life. Venus, Jupiter, Mars and Titan will give interesting weather patterns and sand dunes etc as does the Earth. These are not decreasing entropy states even on a local scale. Powering of life, seems to me, is on a whole different level.

  9. I noticed that in one of Tomkins’ chimp-human papers, he provides the blastN script he used, and he set it to return “ungapped” matches. OF COURSE he can claim that the percent similarity is lower than reported when he will only accept a 100% match in sequence chunks of particular lengths. To wit –


    is 100% similar to:


    but 0% similar to this:


    using an ‘ungapped’ scheme. it strikes me as borderline fraud.

    • That’s not exactly what ungapped does. Ungapped doesn’t use the gapopen and gapextend parameters, which means if there is a single mutation in the sequence, it won’t be able to continue the alignment. Ungapped is a good way of getting an idea of the substitution rate excluding indels. Tomkins should not be using ungapped. Blast will return 89% in your example, with a length of 9 bases.


Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s