"undefined column selected" with pglmm_plot_ranef

[jusinowicz]

Hello,

First, thanks for this awesome R package! It's allowing me to do things that I haven't been able to do with any other GLMM package.

I'm trying to use pglmm_plot_ranef() to extract the random effects predictions on simulated data and I noticed that if I name my "species" phylogeny variable anything other than "sp" that it returns the error:

Error in [.data.frame(data, , sp.var) : undefined columns selected

I ran into this problem with a simulated data set, but it seems to be reproducible running the "### Detailed analysis showing covariance matrices ###" example under the pglmm() documentation. If I run this simulation but change the name of "sp" in the data frame "d" to anything else (and then change the name in the calls to pglmm and pglmm_plot_ranef, of course), then pglmm_plot_ranef will return the error (even though pglmm runs fine). E.g. at these lines change the name and run:

d <- data.frame(species= rep(phy$tip.label, times = nsite),
site = rep(1:nsite, each = nspp))

d$Y <- Y.sp + Y.site + Y.attract + Y.e

#Works
pglmm(Y ~ 1 + (1|species__) + (1|site) + (1|species__@site), data = d, cov_ranef = list(species = phy))

#Does not
re <- pglmm_plot_ranef(Y ~ 1 + (1|species__) + (1|site) + (1|species__@site), data = d,
cov_ranef = list(species= phy), show.image = FALSE, show.sim.image = FALSE)

Am I missing something?
Thanks.

[daijiang]

Hi Jacob!
Can you try pglmm_plot_ranef(sp.var = "species") and see whether that work?

Btw, it has been several years since our last interaction! Hope to catch up with you sometime soon.

Best,
Daijiang

[jusinowicz]

That worked! Was that in the documentation somewhere and I missed it? Sorry if it was. It has been awhile! It's good to hear from you. Thanks for the fast reply! Best, Jacob

…

On Fri, Mar 26, 2021 at 10:37 AM Daijiang Li ***@***.***> wrote: Hi Jacob! Can you try pglmm_plot_ranef(sp.var = "species") and see whether that work? Btw, it has been several years since our last interaction! Hope to catch up with you sometime soon. Best, Daijiang — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#65 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AHOVEWXSNPBILSA46IBRONTTFTA4ZANCNFSM4Z3ZRP5Q> .