GitHub
10/13/2026, 6:32 AMdexseq/dtu module in nf-core/modules#13138, which tests differential transcript usage with DEXSeq.
The modules branch has no replicated transcript counts with multi-isoform genes: the yeast Salmon and kallisto results only contain single-transcript genes, and the human and mouse expression tables are gene level. The module test currently builds its input from the rnasplice branch, which a reviewer flagged as fragile.
New files in `data/delete_me/dexseq/dtu/`:
• chrX.transcript_counts.tsv (67 KB): Salmon transcript counts, imported with tximport, of two GBR and two YRI Geuvadis samples on chromosome X (from testdata/salmon_quant/salmon_quant.tar.gz and reference/genes_chrX.gtf of the rnasplice branch). Only transcripts with a non-zero count in at least one sample are kept, and only genes with at least two of these transcripts: 1,389 transcripts in 365 genes.
• `chrX.samplesheet.csv`: sample IDs and condition (GBR or YRI).
• `README.md`: how the files were made.
Generated by Claude Opus 5.5
PR checklist
Needed by nf-core/modules#13138 (dexseq/dtu).
• I read the test-data specifications
• I tried testing my module with available data and there is no compatible data in test-datasets
• I made sure to submit the smallest dataset possible (e.g. try .tar.gz for larger directories)
🤖 Generated with Claude Code
nf-core/test-datasetsGitHub
10/13/2026, 6:43 AMGitHub
10/13/2026, 6:44 AM