bio-multi-omics-mgwas-integration

What problem does it solve?

Enables researchers to connect genetic variation to metabolite levels and test causal relationships, removing the manual burden of assembling GWAS-ready phenotypes, running association scans, and performing downstream colocalization and Mendelian randomization analyses.

Core Features & Use Cases

• End-to-end mGWAS workflow: guidance for phenotype transformation, genotype quality control, PLINK association testing, and clumping to identify independent mQTLs.
• Locus-level inference: procedures for extracting regional summary statistics and running Bayesian colocalization with coloc to assess shared causal variants with disease traits.
• Causal inference: instrument selection, harmonization, and TwoSampleMR workflows for estimating causal effects of metabolites on disease outcomes.
• Use Case: Convert metabolite concentration matrices into PLINK phenotypes, run per-metabolite GWAS, clump significant hits, colocalize with a disease GWAS, and perform two-sample MR to assess causality.

Quick Start

Run a PLINK association for a transformed metabolite phenotype with appropriate covariates, clump genome-wide significant hits, and then run coloc and TwoSampleMR in R to test colocalization and metabolite-to-disease causality.