Gimme maelstrom didn't work with custom pfm file

[Francis3209]

Describe the bug
Hi Simon,
when I run Gimme maelstrom with my custom .pfm file , an error appear at the beginning of the analysis. Strangely, the standard gimme motifs analysis with the same .pfm didn't give any problem after several runs.

To Reproduce
gimme maelstrom -p ${PFM_FILE} ${INPUT_TABLE} ${GENOME_PATH} ${OUTPUT_DIR}
rm -r /.../tempfiles/cache/gimmemotifs

Expected behavior

Error logs
The following log messages appear:
2024-03-27 10:18:11,629 - INFO - Starting maelstrom
2024-03-27 10:18:11,651 - INFO - Input is not mean-centered, setting the mean of all rows to 0.
2024-03-27 10:18:11,652 - INFO - Use --nocenter if you know what you're doing and want to change this behavior.
2024-03-27 10:18:11,652 - INFO - Note that if you use count data (ChIP-seq, ATAC-seq) we recommend to first transform your data, for instance using log2(), and to normalize between samples. To create a table suitable for maelstrom you can use the coverage_table script included with GimmeMotifs.
2024-03-27 10:18:11,921 - INFO - motif scanning (counts)
2024-03-27 10:18:11,921 - INFO - reading table
2024-03-27 10:18:35,794 - WARNING - multiple motifs with same id: ATF1_HUMAN.H11MO.0.B
2024-03-27 10:18:44,367 - INFO - using 14000 sequences
2024-03-27 10:18:44,404 - INFO - Creating index for genomic GC frequencies.
2024-03-27 10:27:06,498 - INFO - setting threshold
2024-03-27 10:27:06,608 - WARNING - multiple motifs with same id: ATF1_HUMAN.H11MO.0.B
Traceback (most recent call last):
File "/hpcnfs/data/GN2/fgandolfi/.conda/envs/gimmeMotifsEnv/bin/gimme", line 12, in
cli(sys.argv[1:])
File "/hpcnfs/data/GN2/fgandolfi/.conda/envs/gimmeMotifsEnv/lib/python3.10/site-packages/gimmemotifs/cli.py", line 755, in cli
args.func(args)
File "/hpcnfs/data/GN2/fgandolfi/.conda/envs/gimmeMotifsEnv/lib/python3.10/site-packages/gimmemotifs/commands/maelstrom.py", line 42, in maelstrom
run_maelstrom(
File "/hpcnfs/data/GN2/fgandolfi/.conda/envs/gimmeMotifsEnv/lib/python3.10/site-packages/gimmemotifs/maelstrom/init.py", line 192, in run_maelstrom
counts = scan_regionfile_to_table(
File "/hpcnfs/data/GN2/fgandolfi/.conda/envs/gimmeMotifsEnv/lib/python3.10/site-packages/gimmemotifs/scanner/init.py", line 154, in scan_regionfile_to_table
s.set_threshold(fpr=FPR)
File "/hpcnfs/data/GN2/fgandolfi/.conda/envs/gimmeMotifsEnv/lib/python3.10/site-packages/gimmemotifs/scanner/base.py", line 438, in set_threshold
cache.set(k, df[motif.id])
File "/hpcnfs/data/GN2/fgandolfi/.conda/envs/gimmeMotifsEnv/lib/python3.10/site-packages/diskcache/core.py", line 796, in set
with self._transact(retry, filename) as (sql, cleanup):
File "/hpcnfs/data/GN2/fgandolfi/.conda/envs/gimmeMotifsEnv/lib/python3.10/contextlib.py", line 135, in enter
return next(self.gen)
File "/hpcnfs/data/GN2/fgandolfi/.conda/envs/gimmeMotifsEnv/lib/python3.10/site-packages/diskcache/core.py", line 730, in _transact
raise Timeout from None
diskcache.core.Timeout

Installation information (please complete the following information):

• OS: Linux CentOS 7
• Installation Conda Environment
• Version 4.9.2

Additional context
The gimme maelstrom analysis works fine when the custom .pfm file is not provided (i.e.: the default gimme motif database of pwms is used), so I guess some issue related to the custom pfm used

Thanks a lot in advance

Francesco