<#2215 Add RPL28 eukaryotic test data for MetaEuk ...
# test-data
g
#2215 Add RPL28 eukaryotic test data for MetaEuk ORF calling (nf-core/metatdenovo #459) Pull request opened by erikrikarddaniel Description Adds a small eukaryotic test dataset for nf-core/metatdenovo's upcoming MetaEuk ORF-caller support (nf-core/metatdenovo#459), and the two related consolidation issues it's designed to also cover (#460, #463). metatdenovo's existing test data has no eukaryotic genomic (intron-containing) content, so it can't exercise splice-aware gene calling or cross-caller/cross-contig ORF consolidation. Source: the Saccharomyces cerevisiae S288C RPL28 locus (60S ribosomal protein uL15), real NCBI sequence, has a single well-annotated intron. Three read sets simulated from it with
wgsim
(genomic, spliced mRNA, and a synonymously-recoded variant of the same CDS) are engineered so that co-assembly with MEGAHIT produces two verified scenarios: a same-contig, splice-aware two-exon gene call, and a cross-contig, protein-identical pair of ORFs. Full construction details and exact assembly outcomes are in the branch's own
README.md
under "Datasets for nf-core/metatdenovo". New files, all under `test_data/metaeuk/`: •
rpl28_genomic_R1/R2.fastq.gz
•
rpl28_mrna_R1/R2.fastq.gz
•
rpl28_mrna_recoded_R1/R2.fastq.gz
•
rpl28_reference_db.faa
Plus
samplesheet/samplesheet_metaeuk.csv
, referencing the three read sets as three samples. Already verified end-to-end against a real nf-core/metatdenovo pipeline run (
-profile test_metaeuk
, MetaEuk
--orf_caller
): assembles to 2 contigs as designed, MetaEuk calls the expected two-exon splice-aware gene, featureCounts sums both exon segments into one gene-level count. Generated by Claude, reviewed by @Daniel Lundin before opening. nf-core/test-datasets
Review on #2215 Add RPL28 eukaryotic test data for MetaEuk ORF calling (nf-core/metatdenovo #459) Review approved by d4straub on erikrikarddaniel's pull request Nice docu, tiny fastq files! LGTM nf-core/test-datasets