[R-sig-phylo] Show Informative Sites?

Hi all, A simple question to which I can't seem to find an answer: is there a way to show only the informative sites of a DNA sequence dataset in R? Thanks, Jimmy -- Jimmy O'Donnell PhD Candidate Ecology and Evolutionary Biology University of California Santa Cruz, CA 95060

Re: [R-sig-phylo] Show Informative Sites?

All sites are informative under likelihood, but I assume you mean parsimony-informative, in which case all you have to do is count which sites are either (a) uniform or (b) uniform except for differences found only in a single species. Probably easiest if you convert the read.nexus.data

Re: [R-sig-phylo] Show Informative Sites?

Nick Matzke wrote: All sites are informative under likelihood, ... and under Bayesian methods, and under distance matrix methods too. If you leave out uninformative sites without compensating for doing that by an explicit statistical correction, you will have some very unpleasant surprises.