Hi Chris.
For part (1) if you want Robinson-Foulds distance (aka. symmetric
distance) you could use multiRF in phytools
(http://blog.phytools.org/2013/09/getting-robinson-foulds-distances-for.html)
or RF.dist in phangorn, which I believe can now handle multiple trees &
is probably significantly faster. You could also try
Rtreedist(...,method="symmetric") from Rphylip, which requires that you
first install Felsenstein's PHYLIP package.
For part (2) there are a variety of options for batching model-based
statistical phylogeny estimation across a set of trees. If you want to
do it from within R, just create a list of your alignments (objects of
class DNAbin). Then you could, for instance, do the following using
phangorn:
foo<-function(x){
data<-as.phyDat(x)
obj<-pml(rtree(n=length(data),tip.label=names(data)),data,k=4)
fit<-optim.pml(obj,optNni=TRUE,opbBf=TRUE,optQ=TRUE,optGamma=TRUE,optEdge=TRUE)
fit$tree
}
MLtrees<-lapply(X,foo)
in which X is your list of alignments. (This is just an example & you
probably need to tweak this as it is untested.)
You could also call DNAML in the PHYLIP package using Rphylip. In this
case it would look something like this:
MLtrees<-lapply(X,Rdnaml,speedier=TRUE,global=FALSE) # for example
although in this case the model parameters are not optimized.
Finally, it is possible to call RAxML and mrbayes from R using phyloch.
You will have to consult the documentation pages of phyloch
(http://www.christophheibl.de/Rpackages.html) to see how to do that, or
you could refer to my very basic tutorial here:
http://www.phytools.org/mpma/Exercise_7.1/
Good luck.
All the best, Liam
Liam J. Revell, Assistant Professor of Biology
University of Massachusetts Boston
web: http://faculty.umb.edu/liam.revell/
email: liam.rev...@umb.edu
blog: http://blog.phytools.org
On 4/14/2014 8:13 AM, Chris Buddenhagen wrote:
I would like to make pairwise comparisons of topological similarity between
all possible combinations of 518 gene trees.
The expected output would be a matrix of topological distances for each
gene tree to each other tree.
Any suggestions?
Also as an aside, is there a way to mechanize the estimation of 100s of
gene trees from alignments, such that the best model models of nucleotide
substitution is chosen objectively and then a tree is generated using
likelihood or Bayesian methods. Ideally I give the program the folder of
100s of alignments and tell it to go and wait for the gene trees.
Best
Chris Buddenhagen
Florida State University
cbuddenha...@gmail.com
[[alternative HTML version deleted]]
_______________________________________________
R-sig-phylo mailing list - R-sig-phylo@r-project.org
https://stat.ethz.ch/mailman/listinfo/r-sig-phylo
Searchable archive at http://www.mail-archive.com/r-sig-phylo@r-project.org/
_______________________________________________
R-sig-phylo mailing list - R-sig-phylo@r-project.org
https://stat.ethz.ch/mailman/listinfo/r-sig-phylo
Searchable archive at http://www.mail-archive.com/r-sig-phylo@r-project.org/
