Saturday, July 7, 2012

Evolution2012: Saturday

Greetings from Ottawa!  I'm here for the annual evolution conference, which has been pretty fun so far.  I know some readers might find it hard to believe that a creationist would enjoy a big conference with hundreds of evolutionary biologists, but I am.  True, I'm not exactly "one of the boys," but I'm learning a lot, and I'm really enjoying Ottawa.

For me one of the biggest delights of the conference are the talks I didn't know I would like.  I typically read through the program and pick out sessions I think I'll be interested in, but inevitably there are talks with cryptic titles of little interest to me that turn out to be great.  This morning, Graham Slater of UCLA surprised me with his talk "Robust regression and posterior predictive approaches to improve the fit of evolutionary models for quantitative traits."  I confess that that title did not appeal to me, but it was well worth it.

Slater is attempting to evaluate "early burst" models of evolution, George Gaylord Simpson's idea of a paleontological adaptive radiation (not to be confused with the way an ecologist might use the term).  An early burst is the idea that phenotypic evolution is very rapid early in the diversification of some lineages.  For example, the burst of mammalian diversity just after the dinosaurs died outt (according to conventional evolution, of course).  When I was at Evolution2009 in Idaho, I reported on Luke Harmon's presentation on the same subject, where he argued that early bursts are rare.  This work was later published in Evolution.  Slater is using simulations to evaluate our ability to detect early bursts, and he argued that they might not be as rare as Harmon et al. concluded.

Why is that such a cool talk?  Perhaps you've heard the creationist argument about disparity and diversity in the fossil record?  The argument goes something like this:  Evolution predicts the gradual accumulation of differences that eventually lead to disparate organisms, but the fossil record shows disparity from the start with gradual accumulation of variations of each disparate type.  (I don't mean to pick on anybody in particular, but here's a vesion of that very argument.)  To my ears, that sounds like creationsits are claiming that "early bursts" should be common.  I hesitate to equate the two claims directly because creationists would want to qualify their claim to account for Flood vs. post-Flood deposition and created kind diversification, but let's say for the sake of argument that creationists think early bursts should be common.  Harmon et al.'s results suggest otherwise: early bursts are rare.  Since Harmon et al. used some molecular phylogenetic methods rather than just assessing the fossil record directly, I don't think that Harmon et al. have falsified the creationist claim, but their results certainly give me pause about claiming that disparity precedes diversity.  At the very least I think their results call for a more careful re-evaluation of the creationist argument.

Now Slater's results suggest that Harmon et al. might have missed some early bursts, which I think is an important finding.  He hasn't restored my full confidence in the disparity precedes diversity claim, but his work has renewed my interest in seeing a more careful evaluation of the fossil evidence of disparity and diversity.

I also heard a great talk on the Type 3 Secretion System by Sophie Abby of the Institut Pasteur.  She argued (among other things) that the T3SS was derived from the bacterial flagellum rather than the other way around.  I thought that might be of interest to you irreducible complexity fans.  It was really interesting, and I'm looking forward to the published paper.

So that was day 1.

Feedback? Email me at toddcharleswood [at] gmail [dot] com.