Tuesday, April 20, 2010
Huelsenbeck 2003
Huelsenbeck et al. (2003) write that "the parsimony method has been the only method available for mapping characters onto a phylogenetic tree" (Discussion, p. 147). They then go on to cite many of the Schluter, Pagel etc. papers that have used Markov-chain methods in a maximum-likelihood context to reconstruct ancestral states and examine character correlation. At the time of publication (2003), were ML methods not already being used for mapping morphological characters? It seems to me that now, folks favor standard ML methods for character mapping over parsimony (that much is clear), but I have seen Huelsenbeck's et al. stochastic Bayesian approach less frequently. So, I am confused as to where this paper fits into the 'history' of development of phylogenetic analysis. The authors push it as a novel method for mapping morphological characters, but retrospectively it appears to me that the novel contribution of this paper is the ability to map changes along the branches of the tree, rather than the use of Markov models to examine morphological or other character evolution on a tree per se.
Subscribe to:
Post Comments (Atom)
I found that statement confusing as well, because there are clearly papers from the 1990's which use likelihood-based methods of trait reconstruction, including the Pagel 1999 paper we read. I agree that Pagel 1999 does not seem to map changes within the branches. I wonder how much difference that actually makes though. I think likelihood-based methods of character evolution generally incorporate reversals along a branch even if where the changes occur along the branch isn't mapped out. The Huelsenbeck paper does map this, but where exactly the change occurs may differ a lot between realizations of simulated character histories with the observed character values of the terminal taxa to inferring their character histories rather than directly computing the likelihood. I wonder how much confidence you would be able to place in a change occurring in a particular place along a branch anyway. Maybe it affects estimates of parameters related to character evolution? I thought the most important aspect of this paper which differed from at least Pagel 1999 and some of the Schulter papers from the 1990's was the ability to incorporate phylogenetic uncertainity in ancestral state reconstruction. I thought that performing simulations along the Bayesian posterior probability distributions was a more elegant way of doing this than all the complications introduced in the MacClade manual for dealing with polytomies and multiple most parsimonious trees. I'm not sure what attempts were made to do this before within the likelihood framework.
ReplyDelete