nf-core/metaboigniter - Updates and benchmarking

Edit

Goal

Metaboigniter updates and evaluation of adding mzmine as a backend for computations.

Description

nf-core/metaboigniter is nf-core’s pipeline for pre-processing mass-spectrometry metabolomics data. The last release, 2.0.1, is now over two years old. The pipeline still seems to be used, but not maintained.

MzMine is a popular open-source alternative to the current OpenMS backend with a possibility of batch processing. In principle users should be able to choose a backend and also benchmark different backends against each other for their applications. Adding common benchmark datasets is something that is still missing for metaboigniter as far as I can tell.

Tasks

We’ll start by discussing the most urgent pipeline updates and see which expertise on Metabolomics data processing we have in the room.

Open Current issues:

• sample-order mismatch between the ConsensusXML and quantification table (#113, Jan 2026),

• add a step to ensure indexed-mzML files (#95, open since May 2024),

• check on updates for PYOPENMS_MSMAPPING module (#103, Dec 2024).

• update to newer OpenMS releases

Benchmarking MzMine on published benchmarking datasets:

• discuss pitfalls of benchmarking metabolomics data
• get an idea of good parameters for MzMine and OpenMS for benchmarking

Docker image and template updates:

• OpenMS image is using OpenMS 3.0.0, where we by now are at version 3.5.0
• local models use python scripts from bin folder
• merge 3 PRs with updates