Hi Agus (& list).
Thanks for sending along your files. This is what I discovered:
1) Your tree had an extra taxon ("robusta") not represented in your data.
2) However, when removed the analysis still doesn't work because some of
your environmental variables seem to be perfect linear combinations of
other variables (this creates a "computationally singular" error).
3) When I subsampled the columns of X, the function returned the
following error:
Error in ccs > 0 : invalid comparison with complex values
This turns out to be because some parts of the calculation result in
complex numbers. (I think this is just rounding error in spectral
decomposition because some eigenvalues of the cross-product matrices are
almost exactly zero and have tiny imaginary parts.) However in later
steps the imaginary part of these numbers go to zero. R doesn't
automatically change the class of a complex number to real when it's
imaginary part disappears.
I'll try and fix this final issue, but I can't do much about (1) & (2).
Thanks for the report!
- 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 7/17/2013 2:10 AM, Agus Camacho wrote:
Dear list,
Im trying to perform a phylogenentic CCA, using the function phyl.cca from
Liam Revell's site.
I used several steps to guarantee that the names of the compared matrices
(enviro1 and morpho1) are the same, in the same order, the same as the tree,
etc.
However it keeps sending me the error message:
Error in phyl.cca(tree, morpho, enviro1) : subscript out of bounds
It smells to something is not pairing well among the two matrices, but
i cant tell what it could be.
Any hint?
The whole script im using is:
enviro1=na.omit(enviro[tree$tip,c(20:40)])> filter=rownames(enviro1)>
enviro1=as.matrix(enviro1)> morpho1=na.omit(morpho[tree$tip,c(3:5)])>
morpho1=morpho1[filter,]> morpho1=as.matrix(morpho1)> dim(enviro1)[1] 62 21> dim(morpho1)[1]
62 3> setdiff(rownames(morpho1),rownames(enviro1))character(0)
result<-phyl.cca(tree,morpho,enviro1)Error in phyl.cca(tree, morpho, enviro1) :
subscript out of bounds
_______________________________________________
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/
