Gene Hunt said:
Brian Sidlauskas did this in an Evolution paper in 2008 (based on
Yang and Rannala 1997):
Sidlauskas, Brian L. 2008. Continuous and arrested morphological
diversification in sister clades of characiform fishes: a
phylomorphospace approach. Evolution 62(12): 3135-3156.
He had a morphological tree without fossil or molecular branch
lengths. The branch lengths were based on rates of speciation,
extinction and sampling, and so it would be a different approach
than the coalescent theory, I believe. I think Brian has R code
for it; you may want to contact him directly.
In my 2004 book, on pages 570-571 I note the use of a time
transformation that makes simulation of birth-death processes easy.
This is based on Bruce Rannala's 1997 work.
If you want to start out with the first fork in the tree at time T
ago, you want to take your N tip species and choose uniformly from
2, 3, ..., N-1 to choose the number of those tips that are descended
from the left branch of the fork. Say this is called K. Then you
need to simulate a birth-death process for K tips (the subtree
descended from the left fork), and another for N-K tips for the right
fork. The result will be a fairly quick algorithm for simulating a
BD process that has its oldest fork T units of time ago and leads to
exactly N tips.
If instead you just want the process to start T units of time ago,
and have the oldest fork occur some time after that (determined by
the BD process), then that is even easier using the time-
transformation trick.
I do not know the R packages well enough to know whether one of them
implements this, but I am sure someone will comment on that.
Joe
----
Joe Felsenstein, j...@gs.washington.edu
Dept. of Genome Sciences, Univ. of Washington
Box 355065, Seattle, WA 98195-5065 USA
_______________________________________________
R-sig-phylo mailing list
R-sig-phylo@r-project.org
https://stat.ethz.ch/mailman/listinfo/r-sig-phylo
