gm.prcomp issue?

Question

gm.prcomp issue?

Closed this issue 3 years ago · 5 comments

Please, how to visualize Phylomorphospace in ver. 4.0 Geomorph?
I tried
plot(gm.prcomp(Y.gpa$coords, phy=mytree, align.to.phy = TRUE))
and got this:
Error in h(simpleError(msg, call)) :
error in evaluating the argument 'x' in selecting a method for function 'as.matrix': subscript out of bounds

"plot(gm.prcomp(Y.gpa$coords)" works well while standing alone without the tree data.
Thanks!

Answer 1 · 2021-06-22T05:22:35.000Z

Have you installed the latest version of RRPP? If this is not the issue, I´m afraid this is difficult to diagnose without the data, if you want to send me a bin with your Y.gpa object I can have a look.

Answer 2 · 2021-09-03T00:25:36.000Z

Please, is it possible to get any full script for 4.0 Geomorph version so I could fully follow it (I mean from uploading to plotting)? I tried to use manual but there are some gaps in explanation or what type of data is used. For example "phy" data is included already in it. I have one script from friend and his analysis, but it is for 3.0 version and some things just does not work anymore (f.e. phylomorphospace, which is now integrated into gm.prcomp but it doesn't want to work and few other issues. Thanks a lot!

Answer 3 · 2021-09-03T00:38:43.000Z

There is a vignette just for this purpose.

…

Sent from my iPhone

On Sep 2, 2021, at 8:25 PM, stanleewild ***@***.***> wrote: Please, is it possible to get any full script for 4.0 Geomorph version so I could fully follow it (I mean from uploading to plotting)? I tried to use manual but there are some gaps in explanation or what type of data is used. For example "phy" data is included already in it. I have one script from friend and his analysis, but it is for 3.0 version and some things just does not work anymore (f.e. phylomorphospace, which is now integrated into gm.prcomp but it doesn't want to work and few other issues. Thanks a lot! — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub, or unsubscribe. Triage notifications on the go with GitHub Mobile for iOS or Android.

Answer 4 · 2021-09-03T00:49:33.000Z

Thanks. But for example in "Y.gpa <- gpagen(plethspecies$land)", the file "plethspecies" contains already both landmark and phy data. So later if phy data is needed there is only "plethspecies$phy" used. I have my data separated and I don't know how to merge it in this way. When I try to use only my "phy" (meaning tree data, lets say "mytree"), it does not work. It is pruned and in older version it worked.
PCA.w.phylo <- gm.prcomp(Y.gpa$coords, phy = plethspecies$phy) #your testing data
vs.
PCA.w.phylo <- gm.prcomp(Y.gpa$coords, phy = mytree) # my data
Thank you.

Answer 5 · 2021-09-03T01:28:22.000Z

I’m afraid I do not understand. There is no merging of data. The list, plethspecies is a list, and one can make any kind of list in R. The function is not working because of some linking of different objects in a list. The data and tree can be separate. You have said that the function does not work but there can be a multitude of reasons that it might not work. You have not given any indication of how it does not work. The only recommendation I can give is to check out geomorph.data.frame for making a list (“merging”). I suspect that your pruned tree has a different number of taxa then your data. Only a guess without an example.

…

Sent from my iPhone

On Sep 2, 2021, at 8:49 PM, stanleewild ***@***.***> wrote: Thanks. But for example in "Y.gpa <- gpagen(plethspecies$land)", the file "plethspecies" contains already both landmark and phy data. So later if phy data is needed there is only "plethspecies$phy" used. I have my data separated and I don't know how to merge it in this way. When I try to use only my "phy" (meaning tree data, lets say "mytree"), it does not work. It is pruned and in older version it worked. PCA.w.phylo <- gm.prcomp(Y.gpa$coords, phy = plethspecies$phy) #your testing data vs. PCA.w.phylo <- gm.prcomp(Y.gpa$coords, phy = mytree) # my data Thank you. — You are receiving this because you commented. Reply to this email directly, view it on GitHub, or unsubscribe. Triage notifications on the go with GitHub Mobile for iOS or Android.