bcftools: TypeError

[ewels]

Description of bug

From Slack:

---

Hi, I am new to MultiQC. I am running testfull profile on nf-core Sarek pipeline v3.2.3. The whole pipeline (including MultiQC v1.14) worked well before, but I am recently encountering an issue. Is there any possible way to fix this? Sincerely thank you for your support!

File that triggers the error

HCC1395N.tiddit.bcftools_stats.txt

MultiQC Error log

Oops! The 'bcftools' MultiQC module broke... ────────────────╮
  │ Please copy this log and report it at                                        │
  │ https://github.com/ewels/MultiQC/issues                                      │
  │ Please attach a file that triggers the error. The last file found was:       │
  │ ./37/HCC1395N.tiddit.bcftools_stats.txt                                      │
  │                                                                              │
  │ Traceback (most recent call last):                                           │
  │   File "/usr/local/lib/python3.11/site-packages/multiqc/multiqc.py", line 65 │
  │     output = mod()                                                           │
  │              ^^^^^                                                           │
  │   File "/usr/local/lib/python3.11/site-packages/multiqc/modules/bcftools/bcf │
  │     n["stats"] = self.parse_bcftools_stats()                                 │
  │                  ^^^^^^^^^^^^^^^^^^^^^^^^^^^                                 │
  │   File "/usr/local/lib/python3.11/site-packages/multiqc/modules/bcftools/sta │
  │     self.add_data_source(f, s_name, section="stats")                         │
  │   File "/usr/local/lib/python3.11/site-packages/multiqc/modules/base_module. │
  │     source = os.path.abspath(os.path.join(f["root"], f["fn"]))               │
  │                                           ~^^^^^^^^                          │
  │ TypeError: string indices must be integers, not 'str'

Before submitting

• I have read the troubleshooting documentation.
• I am using the latest release of MultiQC.
• I have included a full MultiQC log, not truncated.
• I have attached an input file (.zip if necessary) that triggers the error.

[vladsavelyev]

That's the result of a funny combination of a potential bug in the module, and two bcftools outputs seemingly merged into one file (for HCC1395T_vs_HCC1395N.strelka.somatic_indels.vcf.gz and HCC1395T_vs_HCC1395N.manta.somatic_sv.vcf.gz):

# This file was produced by bcftools stats (1.17+htslib-1.17) and can be plotted using plot-vcfstats.
# The command line was:	bcftools stats  HCC1395T_vs_HCC1395N.strelka.somatic_indels.vcf.gz
#
# Definition of sets:
# ID	[2]id	[3]tab-separated file names
ID	0	HCC1395T_vs_HCC1395N.strelka.somatic_indels.vcf.gz
<...>
# DP	[2]id	[3]bin	[4]number of genotypes	[5]fraction of genotypes (%)	[6]number of sites	[7]fraction of sites (%)

# This file was produced by bcftools stats (1.17+htslib-1.17) and can be plotted using plot-vcfstats.
# The command line was:	bcftools stats  HCC1395T_vs_HCC1395N.manta.somatic_sv.vcf.gz
#
# Definition of sets:
# ID	[2]id	[3]tab-separated file names
ID	0	HCC1395T_vs_HCC1395N.manta.somatic_sv.vcf.gz
<...>

I wonder if it's an expected behaviour? Need to look more into Sarek.

[vladsavelyev]

Ah, figured it's an expected behaviour for bcftools, as I can take multiple VCFs. And the module already supports it, though with a bug. I'll add a fix now and will add the example into the test data.