[R-sig-phylo] Maximum-likelihood tree with continuous characters
Emmanuel.Paradis at mpl.ird.fr
Wed Aug 20 17:05:21 CEST 2008
Le 20.08.2008 13:02, Mark Ruddy a écrit :
> Dear all,
> is there a function/package available for maximum-likelihood tree
> construction from continuous character data in R? I suspect there
> isn't; mlphylo only accepts DNA models.
You're right on both points.
> Briefly, I have a large database of landmark data for geometric
> morphometric analysis. I'm investigating relationships between
> shapes. The only feasible approach to processing the data (bootstrap
> from the original data>GM analysis>build a tree>repeat) is to script.
> I have done this successfully in R with `nj' (neighbour-joining)
> method but would like to try a maximum-likelihood method too.
> Therefore, I suppose my fallback is to script R and PHYLIP's Contml
> through the shell. Does anyone have any other suggestions?
You may script from R itself, writing your data matrix with
write.table(), calling Contml with system(), and reading the tree back
to R with read.tree().
Note that ape has ace() which can compute the likelihood of a single
character but for a given tree only. You can imagine calling it
repeatedly for each character and summing the log-likelihoods, but ape
has no "tree traveller" at the moment (at least in R).
> Many thanks from a new R user,
> Mark Ruddy
> Ph.D. student
> Geography Department
> Royal Holloway
> University of London
> TW20 0EX
> [[alternative HTML version deleted]]
> R-sig-phylo mailing list
> R-sig-phylo at r-project.org
IRD, Montpellier, France
ph: +33 (0)4 67 16 64 47
fax: +33 (0)4 67 16 64 40
More information about the R-sig-phylo