Background: There has been an on-going discussion between Dr. Vincent Torley, a philosopher, and myself on the subject of pseudogenes, synteny, and common descent. These are technical terms and the arguments may seem abstruse, but the topic is important because of its implications. Those who hold to common ancestry accept that, at the very least, each major group of animals is descended from a common ancestor, and some hold that the common ancestry goes all the way back to the first cell. The question is whether we share common ancestry with chickens and opossums, and by implication, with chimpanzees. What we have been discussing for all these posts is this: How strong is the evidence that we have the remnants of an egg yolk protein, vitellogenin, in our genome? Do we come from an egg-laying ancestor?
In a post at Uncommon Descent, “Consider the Opossum,” Vincent Torley took me to task for a number of reasons, but chief among them was that I ignored the evidence for synteny between the opossum and chicken vitellogenin (VTG) genes. (Synteny refers to the alignment of sequences in order and identity between stretches of DNA from two species.) This evidence came from the Brawand paper. I had been meaning to address the opossum data, but events (Torley’s post) got in the way. So rather than debate, I’ll make it simple. The evidence for synteny in opossums and chickens is good. Surprised? I had already planned on saying this before the latest information about human alignments came up.
Now, is this synteny between chicken, opossum, and human evidence for common descent? Could be. Remember, some intelligent design proponents accept common descent based on evidence like this. They may think that this particular scenario happened by purely naturalistic processes. But they also acknowledge that at least some processes could not happen in an unguided fashion. These people may accept common descent but also affirm the necessity and detectability of design. Anyone who affirms the detectability of design is by definition a friend of intelligent design, even if not publicly so.
Others think for various reasons that a designer guided the process, but in such a way that the design is undetectable. They think that the designer has hidden the evidence of design, and may even go so far as to question why he would do such a thing, since everything looks like a purely natural process. This viewpoint might as well be full-blown materialistic neo-Darwinism.
Then there are people who reject common descent to one degree or another. They may proffer different explanations for sequence similarity. Synteny, for example, can be the result of a need for a particular chromosomal organization or structure. Thus, gene expression regulatory blocks can lead to the appearance of conserved synteny. It is also possible that the apparent remains of vitellogenin genes may be being used for other purposes. The vitellogenin protein itself is multi-purpose — it and its derived proteins are involved in the innate immune response in fish. Has anyone looked into this in higher vertebrates? Probably not. There is evidence of long non-coding RNA (lncRNAs) in the region of similarity, plus other possibly functional elements. In response to a comment from Professor Josh Swamidass, I note that lncRNAs can act in cis (on their own region, rather than a distant location) making their location on the chromosome important. It is not therefore beyond the realm of possibility that any similarities we see among opossums, chickens, and humans are there for functional reasons. The signal we see could be the result of essential sequence-dependent function in the region, with different sequences shared between species. I agree with Swamidass that all these proposals are possibilities only and would need to be demonstrated as real and functional.
But even if it’s the case that this region shares the remnants of egg yolk genes and ancient synteny, common descent from an egg-laying ancestor is not the only possible reason there could be old genetic code lying around. It depends on your model for how the designer acts. Consider an analogy with how human programmers work. A programmer I know says that coders routinely use old code in new programs. It’s hard to write new code, but easy to use code that has already been designed, and to plug it into the new program. It takes time to get rid of the unused code that comes along for the ride so to speak. It’s easier to just leave it alone. (My pardon to my friend if I did not express properly what he said.) In fact, in our case, the designer might have plans for that unused code further down the road. I should also say, there is no intent by the programmer to deceive anyone who later comes along to deconstruct the code as to its origin. It looks like ancient code shared between organisms, but it’s not due to common descent. It’s simply the way the designer works.
In all these cases, it could be that once the old code has been created it is subject to the same process of neutral or near-neutral evolution as for standard neo-Darwinian evolution. Mutations happen all the time — most are largely neutral, with little or no effect, so they tend to accumulate. Unless there is some reason to maintain a particular sequence without change (because it has a necessary function), its sequence will tend to drift and degrade, like the vitellogenin genes in opossum, dog, and human are proposed to have done. This will tend to mimic the signal of synteny and pseudogenization in the vitellogenin gene.
This analogy will probably irritate the materialists among us. So let me propose another possibility. The signal we see could result if there is essential sequence-dependent function in the region, with different sequences shared between species.
I acknowledge that common descent is an appealing explanation for this synteny. I know the strength of the common descent argument. I am also aware of its weaknesses, which I will address in my next post. The evidence for common descent is mixed, as we shall see. The reason I doubt common descent is not because of the ways that it succeeds, but rather because of the ways that it fails. And I am still inclined to accept the design explanation because I expect to see function rather than junk in our genomes.
Finally, let me give an answer to one very specific question posed by Vincent Torley, as put to him by Josh Swamidass, as an example of the difference in approach to the data.
Why is the genetic signal stronger in the opossum (a marsupial mammal) than in human beings, who are placentals? The common descent model has an answer to this question: the various lines of mammals made the transition away from eggs at different times, so they inactivated their yolk genes at different times. How does the design model answer this question?
First of all, marsupials are placentals also, but of a different kind. Second, my response to the question is that I hope Torley and Swamidass are not making the mistake of thinking of marsupials as partway up the evolutionary ladder, therefore closer to their egg-laying roots than are eutherian mammals. If common descent is true, the theory says that eutherian mammals and marsupials share a common egg-laying ancestor that diverged from the chicken lineage a long time ago, so present day marsupials and mammals are equidistant from chickens in evolutionary time. There can be no a priori prediction about which should have a vitellogenin pseudogene with closer resemblance to an egg-laying chicken’s gene, without prior, independent knowledge of when egg laying was lost in either lineage.
The Brawand paper is the only paper I could find that assigns a date to when egg yolk proteins were lost in each lineage. It was the reason for their study. They based their estimate on the genetic signal of the degree of degeneracy between the sequences. It is faulty reasoning to say the reason for the greater degeneracy is because of earlier loss of egg-laying, when the reason for the dating of the egg-laying loss is based on said degeneracy.
Common descent cannot explain why egg-laying genes were lost earlier in one lineage than another, since it could have happened either way. Or not at all. See above.
From a design perspective, I would say the reason for the difference in apparent inactivation times is because each animal has a different design. How the DNA is used may differ.
So now on to the next and final explosive post. Stay tuned. It’s my riposte to the “explain and predict” test for common descent versus common design offered by Torley. And I am done with talking about egg yolk proteins.
Photo credit: © ondreicka — stock.adobe.com.