Dechronization Interviews Joe Felsenstein


Dechronization is a group blog devoted to the methodology of phylogenetic tree reconstruction. Today they published an interview with a very influential individual in the field, Joe Felsenstein, whom some of you may recognize as a regular reader of this blog. I’ll quote the first part because it actually pertains to some of the research that I am doing right now—and will talk about this summer in Iowa at SMBE (hopefully) and in Idaho at Evolution.

LH: What are the most exciting recent developments in systematics / comparative methods?

JF: The availability of genome-scale information is certainly one. The arrival of a generation of young researchers who are comfortable with statistical and computational approaches is another. But the most important development is reflected in recent work on coalescent trees of gene copies within trees of species. What this does is tie together between-species molecular evolution and within-species population genetics. Those two lines of work have been developing almost independently since the 1960s. But now, with population samples of sequences at multiple loci in multiple related species, they are coming back together. This is not another Modern Synthesis, but it is a major event that needs a name. How about the “Family Reunion”? Long-estranged relatives who have not been in touch are getting together.

Don’t forget to read the rest.


This comment has been moved to The Bathroom Wall.

Impressive. (And I don’t merely mean the distinguished beard.)

Btw, I find the discussion on practical vs philosophical bayesianism in biology interesting.

I hear, but don’t really know, that bayesian methods are inconsistent with theoretical physics concepts. As I understand it when operating over truly infinite domains (like infinite-dimensional Hilbert spaces of QM) they can use bootstrap priors (or whatever you should call those; pseudo-priors perhaps) and arrive at practical albeit theoretically “unsupported” methods.

But I dunno about the problems of philosophy (!) or priors in general, as long as you test the results.

And that is my point, if I may make one. For my money bayesian methods seems to be great for domain-dependent factual “technical” learning and models (especially unconstrained such), but when you want to pin down parameters and, ultimately, test for potentially domain-independent actual scientific knowledge and theories you are IMVHO moving into frequentist territory however you started out.

Simply because theories and their parameters are testable and likely generalizable outside already studied data, we need those concepts such as practical physical unboundedness and idealized mathematical infinities as mentioned above. (And so I couldn’t care less in which of those categories one philosophically place observable frequencies.)

In other words, I feel a philosophical discussion doesn’t place in methods of testable science. (Or at the very least it is a weak argument.) Because ultimately I sympathize with Joe’s statement that “We are going to have to be able to characterize what we can and can’t know given the data.”

About this Entry

This page contains a single entry by Reed A. Cartwright published on April 23, 2009 5:31 PM.

A Solar System with an Earth-Size World, and a Planet in the Habitable Zone was the previous entry in this blog.

Killing Geology in Florida is the next entry in this blog.

Find recent content on the main index or look in the archives to find all content.



Author Archives

Powered by Movable Type 4.361

Site Meter