[R-sig-phylo] mntd and mpd without community data

Steven Kembel steve.kembel at gmail.com
Thu Mar 17 16:58:48 CET 2011


Hi Scott,
I don't think MPD and MNTD are appropriate measures of tree shape, since factors other than tree shape including the number of tips on the tree and the branch length distribution influence those statistics. I would recommend one of the metrics of tree shape that you mentioned, most of which are implemented in the apTreeshape package.
Best regards,
Steve
On Mar 17, 2011, at 8:45 AM, Scott Chamberlain wrote:

> Hello, 
> 
> 
> I am curious if it is appropriate to calculate mntd (mean nearest taxon distance) and mpd (mean pairwise distance) in the picante package on trees themselves, that is, without community data. 
> 
> We are trying to think of informative metrics that can tell us something about tree shape among lots of different trees. Using mntd and mpd we give the functions a tree and just a vector of all 1's for the community data so that each species is equally abundant. 
> 
> Does this approach make sense? Are there better metrics to use given that we are just dealing with trees without community data? 
> 
> (I am aware of Sackin's, Colless', beta splitting, gamma, etc.)
> 
> 
> 
> Here is a reproducible example of what I am doing:
> require(picante)
> require(ape)
> 
> tree <- rcoal(10)
> abund <- rep(1, 10)
> names(abund) <- tree$tip.label
> mpd_ <- mpd(rbind(abund, abund), # can't have just one vector/community apparently
> cophenetic(tree))[1] # just one of the two numbers needed as they are the same
> 
> 
> 
> Sincerely, 
> Scott Chamberlain
> Rice University, EEB Dept.
> 
> 
> 
> 	[[alternative HTML version deleted]]
> 
> _______________________________________________
> R-sig-phylo mailing list
> R-sig-phylo at r-project.org
> https://stat.ethz.ch/mailman/listinfo/r-sig-phylo



More information about the R-sig-phylo mailing list