Backwards Tree

A new article by Jeffrey Tomkins is out: Epigenetic Study Produces ‘Backwards’ Human-Ape Tree. The subject is a paper called “Regional DNA methylation differences between humans and chimpanzees are associated with genetic changes, transcriptional divergence and disease genes” in the Journal of Human Genetics – it’s open access, but a little dense.

To recap a subject we’ve seen before: various processes, the most famous of which is known as “DNA methylation,” can modify the amount that a gene or other stretch of DNA is expressed and with that its effect. This adds up to what is called epigenetics, a relatively new biological field of investigation. Most epigenetic research has concentrated on what is seen in human cells, but some more recent studies have shown that patterns of methylation are to some extent heritable, and also differ between species. In October of last year Brian Thomas pointed to a study showing differences in methylation in cells in various organs between chimpanzees and humans, and tried to argue that this showed that the two groups could not be related based on the unfounded assumption that methylation patterns could not change – even despite the not-inconsiderable variation observed even within humans.

This study adds gorillas and orangutans into the mix. They included in their paper an illustration of a tree generated from the differences between the patterns observed in each species. Here’s how Tomkins illustrates it (claimed as fair use for criticism) – concentrate first on the tree on the left (his reproduction of the one in the paper) and it’s associated asterixed note:

Tomkins' "backwards tree" figure

It’s not immediately obvious what Tomkins has done, aside from rotating and flipping the diagram, but it looks like he disagrees with the length of some (or at least, one) of the branches. Here’s the original tree:

Methylation tree

The caption in the paper says that, for their figure, “The length of each vertical branch is proportional to the degree of difference in DNA methylation.” Tomkins has apparently extracted the original data, somehow, and calculated a tree with somewhat differing lengths for what are in his figure the horizontal branches. For all his claims of the original authors “mis-representing” their results here, this particular aspect doesn’t seem to be nearly as important to his claim as the overall structure of the tree.

You see, regardless of how long the branches are represented as or what direction the diagram as a whole is oriented, that’s not the commonly-accepted tree made up of humans and our nearest relatives. For one, chimps and humans should be most closely related to each other, not chimps and gorillas. But what then is that tree? If you remember, Tomkins also included above what he thinks it looks like. Go take another look at it – I’ll wait.


That’s not what we’re looking for either, is it? If we again ignore line length and revert the orientation to the vertical, the overall structure of the tree usually presented is this:

Human tree

Humans are most closely related to chimps, and then to gorillas and finally to orangutans. On close inspection, the structure is not actually all that different to the tree calculated, with the only difference being the position of the human branch: in the methylation tree we’re out to the side, while we should be deep within the group.

With the true problem identified, we move to the solution, as seen by the authors of the paper – not that they seem to think of it as a problem:

These results suggest that the methylation status changed more rapidly in the human lineage than in the chimpanzee lineage, consistent with the results of Martin et al. [link]

In other words, while the other tree groups have evolved naturally with respect to the methylation patterns investigated, humans have spun off and done their own thing. There is, however, one more wrinkle – Tomkins says:

Orangutans, who supposedly are the least evolved among apes compared to humans, actually had more DNA methylation patterns similar to humans than chimps or gorillas. And if that was not enough, gorillas were the next closest in similarity to humans with chimps falling out last! According to evolutionary predictions, chimps should have been most similar to humans, then gorillas, and lastly orangutans.

Humans should still be most similar to chimps, but it seems that this is not the case. I suspect that the result may be due to the degree of divergence of the human group coupled with the way the tree was calculated, possibly producing something like this, but I don’t know nearly enough about this subject to say. Do any of you have an idea?

3 thoughts on “Backwards Tree

  1. The problem is that the tree isn’t rooted. If you root it at the point between the orang and where the human branch off, then you get [orang [human [chimp+gorilla]]], which isn’t that different from the “conventional tree” and could then be more easily explained by a faster rate of change in humans versus chimps.

    • From my response to a Facebook friend’s posting of this nonsense:

      The authors (Fukuda et al.) of the research paper “Regional DNA methylation…” on which Tomkins based his specious claims say in the last sentence of the abstract:

      “Our results suggest that DNA methylation changes, often caused by small sequence changes, contribute to transcriptional and phenotypic diversification in hominid evolution.”

      This clearly indicates the authors agree that the findings of their study advance our understanding of evolutionary processes. The research is published in “Nature” which is one of the most, if not the most, prestigious scientific pier review journal. Tomkins apparently finds the research convincing to have based his claim on it. I hope the credibility of the research is apparent.

      Tomkins claims the unrooted phylogenetic tree in figure 2b describes an evolutionary ancestry opposite to current scientific consensus which thereby casts doubt on evolution. The use of an unrooted phylogenetic tree indicates the authors intended to show a relationship between the four species “based on the methylation data” without describing ancestral lineage. Ancestry would have been implied had the authors used a rooted phylogenetic tree. (See figure 2b)

      For me, a person nearly illiterate of biology, to have recognized after some effort the highly educated Tomkins’ “mistake” indictates it was either incompetence or casuistry on his part. Tomkins, being a part of the Institute for Creation Research, is trying incompetently or speciously to make research results fit a prehistoric creation story. Fukuda et al. are trying to make sense of nature using research even if the results lead them away from a preconceived idea. The two are incompatible and it is obvious which approach is superior.

      Here’s a link to the research publication cited by Tomkins so you can judge for yourself.…/v58/n7/full/jhg201355a.html


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

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

Google photo

You are commenting using your Google 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