OmicVerse Bulk Metabolomics — Pathway, Multifactor, DGCA & MOFA

Goal

Cover the four downstream interpretation paths that consume a preprocessed metabolomics AnnData and produce a biological story:

1. Pathway / MSEA — resolve metabolite names to external IDs (HMDB / KEGG / ChEBI / PubChem / LIPID MAPS) and run over-representation (ORA) or GSEA-style ranked enrichment, plot bars / dots.
2. Multi-factor designs — ASCA (ANOVA-SCA), mixed models (MixedLM), MEBA time-series — for studies with treatment × time × patient layouts.
3. Differential correlation (DGCA) — find metabolite pairs whose correlation rewires between conditions, plus per-condition correlation networks.
4. Multi-omics MOFA+ — joint factorization across metabolomics and RNA-seq (or any other AnnData-aligned views).

Plus a worked end-to-end real-data case study: MTBLS1 (urine NMR, Type 2 Diabetes) that ties preprocessing → univariate / multivariate / pathway / multifactor / DGCA / network into one runbook.

Quick Workflow

Pathway / MSEA (t_metabol_03)