Posted 11:17 PM by Robert

Chimp Relations

Scientists reportedly now suspect that chimps and humans are not as closely related as once thought.

Where prior studies suggest that 98.5% of the human genetic code can also be found in the chimp, a new study published in Proceedings of the National Academy of Sciences says the true overlap may be only 95%.

These chimp/human overlap reports have always confused the Man Without Qualities. And it's not just that Jane Goodall's emotional and physical relationships with her chimps have always seemed - for my tastes - a little too close for comfort. (Does she know when she's got hold of an ugly one?). It's just that while it's is all very nice to say that the human and chimp genetic codes have a lot in common, it has always been my understanding that most of the human genetic code is not used at all - but has apparently been deactivated for a very long time. [Indeed, my memory is that in the course of a single, port-drenched dinner two Nobel Prize Winners (one a winner for monoclonal techniques and the other for cat brains) told me that.] Only a small part of the DNA in a human cell is therefore active in cell operations, and I have always assumed the same is true for chimps, too.

My confusion arises because I have never been able to determine from the popular media coverage (such as this linked article) if the "overlap" studies include comparisions of inactive genetic material. If the studies do include such comparisions, then why is it not possible that much of the active human code only overlaps with the inactive chimp code - and vice versa. Wouldn't that mean that a 95% (or 98.5%) "overlap" could be all but meaningless?

Or is the "overlap" implicitly restricted to the active portion of each genome - and that technical aspect of such reports just suppressed in the popular version? I must confess that I have never taken the time and effort to read an original paper. However, I have little doubt that a considerable portion of the readership of the Man Without Qualities curls up with their subscription copy of the Proceedings of the National Academy of Sciences from time to time. So perhaps one of them knows.

Just asking.

UPDATE: TurnedUpToEleven posts a remarkably transparent (given the technical nature of the material) and very enlightening explanation answering the above questions. TurnedUp also makes the telling point that humans appear to be at least as closely related to bonobos as to chimps! Take that, Jane Goodall!

FURTHER UPDATE: By the way, while I very much like TurnedUpToEleven's well-considered response to my post, it is worth noting the portion of that response that actually addresses my questions:

Dr. Britten suggests, based on his analysis of a few million base pairs of the genomes of chimps and humans (about 0.1%), that we may be off in our estimate, and that it is probably closer to 95%. As far as Mr. Musil's concern goes, this study does not distinguish between "useful" DNA and "junk". The best guess, however, is that there isn't going to be a difference, but that, if anything, the functional genes are probably more similar, because they are constrained by their function. That is, the gene that encodes for hemoglobin is more restricted than an equal sized region that does not encode a gene, because what ever changes that hemoglobin gene, its essential function must be preserved. So, it is most likely that the "junk" DNA is not identical, with all the functional genes varying, but rather the other way around. In this sense, the very close similarity between our "junk" and chimp "junk" is further evidence of our close evolutionary pairing.

I think that the "best guess" of a good scientist is worth a lot, and deserves a lot of respect. I also think that a "best guess" is a guess. Further, the proposition that functionality can be expected to constrain genome structure is appealing, as is the corresponding thought that "junk" portions of the genome should be able to drift more.

But if functionality were really so important to genome structure, it is a little odd that there is already a 95% overlap in genome structure even where one considers portions of the genome with no apparent functionality. And if the "junk" portion of the genome actually plays some ill-understood structural role in the genome itself (as TurnedUp suggests may be the case), then how do we know that such internal structural "functionality" does not impose more uniformity than is imposed by external "functionality." For example, if one builds a building out of bricks, the internal engineering funtionality of the walls will have a lot more to say about the ordering of the bricks than does the ultimate external function of the building - a fact which is apparent simply because the exact same building can function as a church or a disco. It seems to me that this kind of thing can only be answered by experiment. But I do not see anything in what has been discussed so far that prohibits a regime in which some simple internal structural role of "junk" DNA may be shared by many animals and cause the "junk" over time to assume a highly unified, definite, shared pattern in many genomes - but where relatively minor differences in external functionality cause vastly greater differences in "useful" DNA sequencing.

In the alternative, if the "junk" DNA represents once-functional DNA which has been superseded by later evolutionary developments, it seems that the "overlap" might be exaggerated in the "junk' portions of the genome. For example, if most of the "junk" comes from a period before chimps and humans evolutionarilly diverged, then all (or most) of the divergence will be reflected only in the "useful' portion of the genome. If that were the case, then studies that do not distinguish between "junk" and "useful" DNA might exaggerate the degree of evolutionary similarity.

Moreover, there are plenty of examples of biological systems that serve the same function but seem to have very different structures. For example, the mouse immune system and the human immune system serve the same "function." But the differences betwen the mouse immune system and the human immune system have demolished more than a few immunological theories. It doesn't seem to me to be much of a stretch to ask whether those differences are reflected at the genome level. Of course, primates have more and deeper "functionality" in common than do rodents and humans.

So, as Aaron Haspel observes, TurnedUp's post is convincing. It convinces me that my first question (Do the studies distinguish between the "junk" and "useful" portions of the genome?) has been definitively answered.(The studies do not distinguish). But my subsequent question (Does this mean the overlap percentage may not be very meaningful?) has not been answered, but does lie within the still-to-be-tested zone of a good scientist's good guess. Actually, several guesses. That's OK with me. Maybe that's what Aaron means by my being "set straight."

Aaron's "mapping fallacy" point is a generalization of the "junk/useful" question. I asked whether the "non-junk" overlap in the chimp and human genones is extensive (Answer: the evidence supports some best guesses that it is extensive - but that's not what these studies are actually saying.). But Aaron points out the danger of concluding too much from even an extensive overlap of the "useful" portions if it is confirmed, a point with which I strongly agree.