Re: Bayesian phylogenetic analysis

Hello,

John Harshman <[EMAIL PROTECTED]> wrote:

> So for any number of taxa, there should be a number of generations that 
> is enough to get a representative sample, swamping any autocorrelation 
> problems, and that number should increase as the number of taxa 
> increases. This would perhaps be more serious if the likelihood peak is 
> broad, or, worse, if there were multiple peaks.

> Thoughts? Experiences?

> (I've been accumulating this impression with runs of 10-20 million 
> generations and data sets around 150 taxa.)

It seems to also depend in the run. Sometimes you find plateau changes
somewhere in between. If the analysis is repeated, burn-in can be shorter
and the plateau is stable. Similar to other analyses, presumably also the
amount of homoplasies may be important. I haven't run analyses with so many
taxa, yet, but possibly choosing the number of generations according to the
number of taxa will not always work. Although, I think, it is a good thing
to start with. It will at least reduce the number of analyses that have to
be repeated.



Best wishes,

Kerstin Hoef-Emden

-- 

Dr. Kerstin Hoef-Emden                  Gyrhofstr. 15
Universität zu Köln                     50931 Köln
Botanisches Institut                    Germany