Bioinformatics QC thresholds should not live in old notebooks

Most bioinformatics teams already generate plenty of QC metrics. The hard part is remembering what those numbers mean months later.

A sample may have a mapping rate, duplication rate, library complexity score, insert size distribution, and a handful of assay-specific checks. The report might look fine in isolation, but the real review question is usually comparative:

• Does this sample look normal for this assay?
• Which previous runs are we comparing against?
• Were these thresholds still current when the report was generated?
• Did this fail because of biology, batch, pipeline version, or assay drift?

Goodomics is a draft attempt at making that context portable. The starting point should stay simple:

goodomics report ./results

That command should produce a useful HTML report from common outputs and custom tables. When a team needs history, the same workflow can ingest results into a local or hosted Goodomics database:

goodomics ingest ./results \
  --project rnaseq-core \
  --report rnaseq-qc@v3 \
  --cohort production-rnaseq-hg38@2026-05

The important part is not a prettier dashboard. The important part is that each QC decision can say: this run was evaluated against this cohort, using this report version and this QC policy version.

That is the small wedge Goodomics is exploring: MultiQC-simple adoption, with cohort-aware context when the team is ready for it.