Error when running cosmos_pkn(organism = "mouse")

[an-altosian]

Dear Omnipath team,

I am trying to build a mouse database for COSMOS. However, I got an error when running the cosmos_pkn function.

[2025-01-06 19:54:22] [SUCCESS] [OmnipathR] Building COSMOS PKN (organism: 10090). This will take 10-30 min at the first time, and will be saved in the cache for later use.
[2025-01-06 19:56:23] [SUCCESS] [OmnipathR] stitch.embl.de: downloaded 39116252 records
[2025-01-06 20:03:01] [SUCCESS] [OmnipathR] stitch.embl.de: downloaded 71116338 records
[2025-01-06 20:03:12] [SUCCESS] [OmnipathR] www.ensembl.org: downloaded 66689 records
[2025-01-06 20:03:23] [SUCCESS] [OmnipathR] www.ensembl.org: downloaded 218929 records
[2025-01-06 20:03:24] [SUCCESS] [OmnipathR] www.ensembl.org: loaded 66689 records from cache
[2025-01-06 20:03:24] [SUCCESS] [OmnipathR] www.ensembl.org: loaded 218929 records from cache
Warning: call dbDisconnect() when finished working with a connection
[2025-01-06 20:04:02] [SUCCESS] [OmnipathR] Chalmers GEM (raw.githubusercontent.com): loaded 1 records from cache
[2025-01-06 20:04:02] [SUCCESS] [OmnipathR] Chalmers GEM (raw.githubusercontent.com): loaded 8454 records from cache
[2025-01-06 20:04:02] [SUCCESS] [OmnipathR] Chalmers GEM (raw.githubusercontent.com): loaded 12987 records from cache
Error in `separate_wider_regex()`:
! Expected each value of `source` to match the pattern, the whole pattern, and nothing but the pattern.
! 14066 values have problems.
ℹ Use `too_few = "debug"` to diagnose the problem.
ℹ Use `too_few = "align_start"` to silence this message.
Backtrace:
  1. OmnipathR::cosmos_pkn(organism = "mouse")
 28. tidyr::separate_wider_regex(...)```

Could you please give me some hints on how to overcome this error? Thanks.

[an-altosian]

After spending some time on it, the error was because this line of code

OmnipathR/R/chalmers_gem.R

Line 359 in 4ffd9ec

patterns = c(id = '(?:MAM|ENSG)\\d+', comp = '[cmxrelgni]?'),

This pattern only works for human because of the "ENSG". For mouse, the chalmers_gem function returned gene symbol instead of ENSEMBL IDs.