Installation failed. version of Rlang and/or miloR?

[RongqinChen]

Dear authors,

Thank you for your great work.

I have tried many versions of Rlang to install the miloDE package, including 4.0.0, 4.1.3, 4.2.0, 4.3.0.
However, I get the same problem when installing miloDE. The error message is

── R CMD build ─────────────────────────────────────────────────────────────────
* checking for file ‘/tmp/RtmpeutvQj/remotes1fa99747f20e5f/MarioniLab-miloDE-3f825b9/DESCRIPTION’ ... OK
* preparing ‘miloDE’:
* checking DESCRIPTION meta-information ... OK
* checking for LF line-endings in source and make files and shell scripts
* checking for empty or unneeded directories
* building ‘miloDE_0.0.0.9000.tar.gz’

Warning message in i.p(...):
“installation of package ‘/tmp/RtmpeutvQj/file1fa997c2ed711/miloDE_0.0.0.9000.tar.gz’ had non-zero exit status”

Does miloDE require specific versions of Rlang and/or miloR? If so, what are the versions?

Thanks

[RongqinChen]

I found some dependencies of miloDE require rlang >= 4.4.0, e.g., ‘MouseGastrulationData’, ‘scWGCNA’, ‘scran’. However, miloR requires rlang< 4.3.0.

[amissarova]

@RongqinChen I think I am a little confused - it seems that the latest version of rlang is 1.1.4: https://cran.r-project.org/web/packages/rlang/index.html

As well as I opened now one of the environments where I know miloDE is installed and I see that rlang version is 1.1.3

Can you please try to install with rlang 1.1.3 or 1.1.4 and let me know if it works?

[RongqinChen]

@amissarova Sorry for the confusion. In fact, I said R as rlang. I have successfully installed miloDE in R4.3.3, however, I am unable to run the calc_AUC_per_neighbourhood function correctly. I have run the following code

stat_auc = calc_AUC_per_neighbourhood(sce_milo , sample_id = ‘sample’ , condition_id = ‘tomato’, min_n_cells_per_sample = 1, BPPARAM = mcparam)

Got the following error:

Error: BiocParallel errors
  4 remote errors, element index: 1, 12, 23, 34
  38 unevaluated and other errors
  first remote error.
Error in MatrixGenerics:::.load_next_suggested_package_to_search(x): Failed to find a rowSds() method for dgTMatrix objects.

Traceback.

1. calc_AUC_per_neighbourhood(sce_milo, sample_id = ‘sample’, condition_id = ‘tomato’, . 
 .     min_n_cells_per_sample = 1, BPPARAM = mcparam)
2. bplapply(colnames(nhoods_sce), function(hood_id) {
 .     out = .get_auc_single_hood(x, nhoods_sce, hood_id, min_cells = 4, . 
 .         min_n_cells_per_sample = min_n_cells_per_sample, n_threads = 1)
 .     return(out)
 . }, BPPARAM = BPPARAM) .
3. bplapply(colnames(nhoods_sce), function(hood_id) {
 .     out = .get_auc_single_hood(x, nhoods_sce, hood_id, min_cells = 4, . 
 .         min_n_cells_per_sample = min_n_cells_per_sample, n_threads = 1)
 .     return(out)
 . }, BPPARAM = BPPARAM) .
4. .bpinit(manager = manager, X = X, FUN = FUN, ARGS = ARGS, BPPARAM = BPPARAM, . 
 .     BPOPTIONS = BPOPTIONS, BPREDO = BPREDO)
``

[amissarova]

Can you please paste here your session_info?
I will look into that - but just also want to say that this step depends on a GitHub package Augur (which is suboptimal for version control) but also is optional - so you can, in principle, proceed without it.

But if you copy-paste your session_info, I will try to look and see why is it failing.

Can I also ask you to try and run it with BPPARAM=NULL and see whether that works?