Use case: repository-scale QC

[bittremieux]

The GNPS, MetaboLights, and Metabolomics Workbench teams have already been computing several metrics for metabolomics data in these repositories.

• Data volume: 700,000–800,000 MS runs from GNPS. MetaboLights, Metabolomics Workbench.
• Metrics: Basic MS1/MS2 scan counts, positive/negative mode, annotations.
• Storage: SQLite and TSV files.
• Action items:
  • Convert metrics to mzQC.
  • Confirm with GNPS/Workbench collaborators about inclusion.
  • Possibly share data via Zenodo.
  • Long term: Provide data through a filterable web interface (external from the repositories).

[bittremieux]

Action item: Idea to use repository-scale QC metrics as a hackathon project. Submit this to the ELIXIR BioHackathon (Berlin; November 3–7, 2025).

• Goal: implement rudimentary mzQC (+ mzTab) export pipelines for data in PRIDE that has SDRF.
• Example for MetaboLights (targeted lipidomics) in Java: Jupyter notebook link
• Example for a MassIVE dataset (single file) in Java: Jupyter notebook link