Friday, April 23, 2010

Selection, Neutrality, and the Appearance of Design

Jerry Coyne has just released his review of two books: The Greatest Show on Earth and What Darwin Got Wrong over at The Nation. Now, a lot of reviews all over the blogosphere have already ripped What Darwin Got Wrong to shreds, and I won't do so again here because I think they have touched all of the bases pretty well. That said, I do have a problem with a phrase that Jerry Coyne made in his blog post (here) linking to his review of the two books. And that is this:


I decided to use the review as a chance to lay out the reasons why biologists accept selection as the only plausible process that produces the appearance of “design” in organisms. (Note to Larry Moran: of course it’s not the only process that causes evolution!)


Now, while Dr. Coyne is acknowledging that other evolutionary processes exist, namely random genetic drift a point often belaboured by Dr. Moran, I take issue with the recurring notion that Selection is the only mechanism capable of producing features that look designed, and I just don't think that is true.


A paper published in 1999 by Arlin  Stoltzfus (who is doing an excellent series of guest posts over at The Sandwalk)called On the Possibility of Constructive Neutral Evolution (here) addresses this. I have mostly been exposed to this topic through my time as a graduate student. Because we study molecular evolution, I think that the idea of Neutral Evolution is much more firmly planted in our minds than it is for many evolutionary biologists, and I have certainly absorbed that paradigm. Neutrality, for me, is the proper default null hypothesis for evolutionary features. Adaptationist explanations require evidence of selection, selection should not be assumed by default.

So what is Constructive Neutral Evolution? Well my primary exposure to this concept, at leats layed out as a package deal, has been talks by Ford Doolittle. For an in depth opinion/review of a recent talk on the subject that Ford gave by an undergraduate student see a post by PsiWavefunction here. In brief Constructive Neutral Evolution lays out how complex features can arise without the complexity of the mechanism being selected for. In the realm of molecular biology there are lots of transient and accidental interactions between molecules.

In brief, Constructive Neutral Evolution and the building of complexity is a ratchet mechanisms. These transient (and unselected for) interactions provide the opportunity for the stabilization of mutations in one or the other binding partner that would otherwise have been deleterious. Under small population sizes these "deleterious" mutations can become fixed and the transient accidental interaction is now required. We have moved from a one component to a two component system, with the two component system now unable to go back. It has moved to be more complex, without positive selection being what drove this increase in complexity, and appearance of design.

I am sure we will see more on this subject over at The Sandwalk. More in depth reading can be done at PsiWavefunction's blog linked above, or even better if you have access read the original paper that I also linked to.

Edit: Some Additional Links:

Rosie Redfield commented on a talk given by Ford Doolittle on the subject here as well that is quite informative and PsiWavefunction has notes posted from the same talk here