Description of the bug

There is a bug in the merging step of hamronization results with the taxonomy results. ARG hits are duplicated in some cases probably because some table joining on the contig names went not correct. The correct hit is always present, but is surrounded by hits of wrong taxa classifications (because wrong contig numbers merged).

This only affects the ARG subworkflow, i.e. the hamronization_complete_summary_taxonomy.tsv.gz file; AMP and BGC subworkflows are fine.

In the below example, the correct contig k141_349539 is assigned to its correct taxid, rank_label etc. but multiple contig IDs starting with the same numbers (but not matching k141_349539) are added to the same ARG hit.

Command used and terminal output

Relevant files

No response

System information

No response