Inconsistent number of columns found

[Hanjiewu]

I upgraded MultiQC to version 1.26 yesterday, but one of my custom reports was skipped, and the message displayed was: “Inconsistent number of columns found in amr_combined.tsv” I’m not sure what went wrong with this particular table, as the other tables had no issues. The code (shown below) wasn’t changed and worked fine with version 1.24. I upgraded to version 1.27 today, but the problem persists.
amr_combined.tsv (2.1 MB)

 amr_results:
        parent_id: 'Other species'
        parent_name: "Other species grouped data"
        parent_description: "Other species grouped data contains MLST, AMR and VFDB sub-sections below it"
        id: 'amr_results'
        section_name: 'amr_results'
        description: "Antimicrobial resistance and Virulence factors prediction"
        plot_type: 'table'

[vlad.savelyev]

Ah, managed to reproduce the issue! Sorry about it. MultiQC strips the trailing spaces, and in your table some rows have empty values in the last column, which leads to the inconsistent number of values after splitting the row.

Fixed this here: Parsing custom content TSV: allow leading empty columns by vladsavelyev · Pull Request #3059 · MultiQC/MultiQC · GitHub

[Hanjiewu]

Thank you for the quick fix. One more thing: what is the limitation on the number of samples allowed for an interactive table? Is there a way to bypass this mechanism?

[ewels (Phil Ewels)]

Yes, using the config option max_table_rows. See the docs for more details:

Configuration | Seqera Docs