Skip to content

fix(fastq_qc_trim_filter_setstrandedness): expose fastp's merged reads - #12895

Merged
pinin4fjords merged 2 commits into
nf-core:masterfrom
pinin4fjords:fix/fastq-qc-trim-filter-fastp-merge
Sep 7, 2026
Merged

fix(fastq_qc_trim_filter_setstrandedness): expose fastp's merged reads#12895
pinin4fjords merged 2 commits into
nf-core:masterfrom
pinin4fjords:fix/fastq-qc-trim-filter-fastp-merge

Conversation

@pinin4fjords

Copy link
Copy Markdown
Member

Summary

FASTQ_FASTQC_UMITOOLS_FASTP already emits fastp's merged overlapping read pairs (FASTP.out.reads_merged), but FASTQ_QC_TRIM_FILTER_SETSTRANDEDNESS never captured or re-emitted that channel. Any caller enabling fastp_merge gets its merged reads silently dropped, with no output, no error, and no way to consume them downstream.

What changed

  • Added ch_trim_reads_merged, captured from FASTQ_FASTQC_UMITOOLS_FASTP.out.trim_reads_merged in the fastp branch, initialised to channel.empty() elsewhere.
  • Emitted it as a new trim_reads_merged subworkflow output.
  • Documented the new output in meta.yml.
  • Added an assertion to the existing homo_sapiens paired-end [fastq] fastp sortmerna test (which already runs with fastp_merge = true) confirming the merged-reads file is non-empty with the expected read count.

This is purely additive — no existing output changes shape, so downstream pipelines that don't consume the new output are unaffected.

Found while reviewing nf-core/riboseq's v2.0.0 release PR, which vendors this subworkflow and hits the bug via --fastp_merge.

@pinin4fjords

Copy link
Copy Markdown
Member Author

Thanks @jonasscheid !

@pinin4fjords
pinin4fjords added this pull request to the merge queue Sep 7, 2026
Merged via the queue into nf-core:master with commit 0853b48 Sep 7, 2026
59 of 61 checks passed
@pinin4fjords
pinin4fjords deleted the fix/fastq-qc-trim-filter-fastp-merge branch September 7, 2026 21:11
pinin4fjords added a commit to nf-core/riboseq that referenced this pull request Sep 8, 2026
…mits

fastq_qc_trim_filter_setstrandedness -> 0853b48c (nf-core/modules#12895)
orftable_fasta_gtf_buildorfcatalogue -> 7baf88e0 (nf-core/modules#12896)

Both PRs are merged. Synced modules.json and the two changed files to
their exact merge-commit content (not current master HEAD, to avoid
pulling in unrelated later changes). orftable_fasta_gtf_buildorfcatalogue
was already byte-identical (riboseq's copy already matched). Resolves
the check_local_copy lint failures that were expected pending these
merges.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants