Grouping structure 'gr' expects only a single grouping term

Interfaces brms
1

Dear Paul,

I have now been working with phylogenetically corrected models in brms for the past two years. I noticed that the previously used cov_ranef function was now replaced with a grouping argument inside the gr function. However, when I tried the implement the same structure as the one I used with cov_ranef, I get the following error: Grouping structure ‘gr’ expects only a single grouping term. I am not well versed enough to understand specifically what it means as I only have one grouping factor in my very simple database. I therefore looked at the example that was posted in the phylogenetic vignette of brms and copied it into my R session. However, I still get the same error when trying to run the model… Would you be able to help me on how to solve this issue?

All the best!

Antoine

Session info:
R version 4.0.0 (2020-04-24)
Platform: x86_64-w64-mingw32/x64 (64-bit)
Running under: Windows 10 x64 (build 18363)

phylo <- ape::read.nexus("https://paul-buerkner.github.io/data/phylo.nex")
data_simple <- read.table(
  "https://paul-buerkner.github.io/data/data_simple.txt", 
  header = TRUE
)

A <- ape::vcv.phylo(phylo)

model_simple <- brm(
  phen ~ cofactor + (1|gr(phylo, cov = A)), 
  data = data_simple, 
  family = gaussian(), 
  data2 = list(A = A),
  prior = c(
    prior(normal(0, 10), "b"),
    prior(normal(0, 50), "Intercept"),
    prior(student_t(3, 0, 20), "sd"),
    prior(student_t(3, 0, 20), "sigma")
  )
)

Output: Error: Grouping structure 'gr' expects only a single grouping term

If you need to use math formula, use Latex syntax:

Y \sim N(\mu, \sigma)

Don’t forget to attach tags (top right of this form) for application area/class of models or other general subject areas your topic touches on.

2

Your brms version is likely just out of date.

3

I am having the same error with my own data and the " Estimating Phylogenetic Multilevel Models with brms" vignette simple data. brms version is 2.13.5

I am running this on our compute cluster here with R version 3.5.1. Could that be contributing to it?

4

I don’t think you have brms 2.13.5 on the system as cov argument exists there already. For what it’s worth, try out also the latest github version of brms.

5

Here is the code and results from our server:

> phylo <- ape::read.nexus("https://paul-buerkner.github.io/data/phylo.nex")
> data_simple <- read.table(
+   "https://paul-buerkner.github.io/data/data_simple.txt",
+   header = TRUE
+ )
> A <- ape::vcv.phylo(phylo)
> model_simple <- brm(
+   phen ~ cofactor + (1|gr(phylo, cov = A)),
+   data = data_simple,
+   family = gaussian(),
+   data2 = list(A = A),
+   prior = c(
+     prior(normal(0, 10), "b"),
+     prior(normal(0, 50), "Intercept"),
+     prior(student_t(3, 0, 20), "sd"),
+     prior(student_t(3, 0, 20), "sigma")
+   )
+ )
Error: Grouping structure 'gr' expects only a single grouping term

 > packageVersion("brms")
[1] ‘2.13.5’


> version
               _
platform       x86_64-pc-linux-gnu
arch           x86_64
os             linux-gnu
system         x86_64, linux-gnu
status
major          3
minor          5.1
year           2018
month          07
day            02
svn rev        74947
language       R
version.string R version 3.5.1 (2018-07-02)
nickname       Feather Spray
6

Can you try again with the brms gihub version?

7

Dear Paul,

I just updated my brms version to 2.13.5 using github and it works perfectly now!

Thank you for your help!

Antoine

8

I am having issues with the devtools package on our cluster, bur installed the github version on my personal machine and the grouping issue was fixed. Thanks Paul!

9

I encounter the same problem on our cluster. brms is version 2.9.0; R version 3.6.1 (2019-07-05).

I am using Miniconda and I don’t think I can install any newer versions of R/brms, but I will find out. Is there a work around for this cluster set up with older versions of brms/R? Should I use cov_ranef instead?

Many thanks,
Annemarie