Skip to content

Metabonaut

Exploring and analyzing LC-MS data — an online tutorial book and end-to-end workflows for LC-MS/MS metabolomics analysis using R/Bioconductor packages. Covers preprocessing, normalization, differential abundance, and annotation; ships a small example LC-MS/MS dataset for reproducible demos.

Talk at EuroBioC2025

  • Speaker: Philippine Louail
  • Day / session: Day 1 (Wed Sep 17, 2025) — morning short-talk session, after the Crowell keynote
  • Talk title: "Metabonaut tutorials for metabolomics workflows"
  • Slides: Zenodo 10.5281/zenodo.17120470
  • Video: not recorded
  • Abstract / schedule: eurobioc2025 schedule

What it does

Metabonaut is curriculum-shaped, not a method package: it walks LC-MS/MS metabolomics analysis end-to-end on the Bioconductor mass-spec stack (xcms, MsExperiment, Spectra, MetaboAnnotation, …) — preprocessing, normalization, differential abundance, and compound annotation — with a small example dataset embedded so each chapter is runnable.

How it works

Core idea. The "method" here is the curriculum architecture itself: chapters are composed so that the output object of one chapter is the input of the next, threading a small example LC-MS/MS dataset through the full Bioconductor mass-spec stack from raw mzML to annotated metabolite list.

Inputs / outputs. Input: a small bundled LC-MS/MS dataset (mzML files) plus a sample metadata table. Output for the learner: a runnable end-to-end notebook covering preprocessing, QC, statistical analysis, and annotation, and the skills to run the same on their own data.

Key innovation. Filling the gap between per-package vignettes and a real-world workflow. Each Bioconductor mass-spec package has its own vignette in isolation — Metabonaut sequences them into one continuous pipeline, including the R↔Python crossover via SpectriPy.

Parameters worth knowing. not specified in vignette; needs talk slides — the book is structured around chapters rather than a parameter surface. Reader-relevant choices are which chapters to skip based on data size and whether to run the SpectriPy/Python-annotation chapter.

Canonical example. The book is organized into eight modules: end-to-end workflow overview; dataset investigation; QC and feature selection (notame); large-scale processing with xcms (>4,000 files); annotation-resource curation; LC-MS/MS annotation via SpectriPy; advanced annotation with RuSirius for formula/structure ID; and dataset alignment for integrating new data with existing references. The small embedded LC-MS/MS dataset carries through every chapter so each step is executable in-place.

Where it fits in the corpus

  • AACR 2026: axis = bioinfo / AI methods; metabolomics is adjacent to cancer-metabolism dossiers
  • notame (entry) — same LC-MS metabolomics audience and complementary roles (Metabonaut = curriculum, notame = workflow library)
  • SpectriPy (entry) — same Bioconductor mass-spec subcommunity (Johannes Rainer's group)
  • ISMB 2026: scaffold
  • GBCC 2025: queued

Notes

Sister to notame in role — curriculum, not a method package. Useful as the "where to send a new lab member" link for anyone starting LC-MS metabolomics on the Bioconductor stack.