--database argument seemingly not working

[nschcolnicov]

Background

When working on the nf-core/smrnaseq pipeline I saw an issue where a user was trying to use a gff file from a database that wasn't any of the ones that mirtop looks for ("miRBase", "MirGeneDB" or "microRNAs"), the database is called RumimiR (http://rumimir.sigenae.org).
nf-core/smrnaseq#329

Expected behavior and actual behavior.

When using the --database argument, I'm still getting the error "Database not found in --mirna rumimir_sheep.gff. Use --database argument to add a custom source."

Steps to reproduce the problem.

Attaching my workdir to help others reproduce the issue, this was all run using this image "community.wave.seqera.io/library/mirtop_pybedtools_pysam_samtools_pruned:60b8208f3dbb2910"
workdir.tar.gz

Here is the command that I used

mirtop \
    gff \
    --database RumimiR \
    --sps oar \
    --hairpin hairpin.fa_igenome.fa_idx.fa \
    --gtf rumimir_sheep.gff \
    -o mirtop \
    pc4.bam

And the error that I'm getting:

/opt/conda/lib/python3.12/site-packages/Bio/pairwise2.py:278: BiopythonDeprecationWarning: Bio.pairwise2 has been deprecated, and we intend to remove it in a future release of Biopython. As an alternative, please consider using Bio.Align.PairwiseAligner as a replacement, and contact the Biopython developers if you still need the Bio.pairwise2 module.
  warnings.warn(
10/02/2024 11:13:01 INFO Run annotation
10/02/2024 11:13:01 ERROR Database not found in --mirna rumimir_sheep.gff. Use --database argument to add a custom source.
Traceback (most recent call last):
  File "/opt/conda/bin/mirtop", line 10, in <module>
    sys.exit(main())
             ^^^^^^
  File "/opt/conda/lib/python3.12/site-packages/mirtop/command_line.py", line 31, in main
    reader(kwargs["args"])
  File "/opt/conda/lib/python3.12/site-packages/mirtop/gff/__init__.py", line 28, in reader
    matures = mapper.read_gtf_to_precursor(args.gtf)
              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/opt/conda/lib/python3.12/site-packages/mirtop/mirna/mapper.py", line 164, in read_gtf_to_precursor
    if _guess_database_file(gtf).find("miRBase") > -1:
       ^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/opt/conda/lib/python3.12/site-packages/mirtop/mirna/mapper.py", line 40, in _guess_database_file
    raise ValueError("Database not found in %s header" % gff)
ValueError: Database not found in rumimir_sheep.gff header

FYI @lpantano

[atrigila]

@nschcolnicov I think you might need to pass the --database RumimiR-Dec2022 instead of --database RumimiR as that seems to be the identifier in the file:

1	RumimiR-Dec2022	miRNA	89245	89266	.	-	.	Specie=Caprine; Name=chr1_811; RumimiRID=Rum-chi-00001; Tissue=Mammary_tissue

[nschcolnicov]

Hi @atrigila I tried using that value and I got the same error. Also, I see that the mirtop script extracts the database from the header instead of from the body, which is why I'm using that value:

mirtop/mirtop/mirna/mapper.py

Line 29 in 603f45c

if not line.startswith("#"):

# Provided by RumimiR database (v. Dec2022) available at http://rumimir.sigenae.org
# File contains 22348 miRNA