The Challenge

MISAME‑III aims to evaluate a nutritional supplement for pregnant and lactating women and to understand the biology underlying fetal and infant health in rural Burkina Faso. The project involves multiple omics—including metabolomics, proteomics, metagenomics and methylomics—and requires sophisticated analysis to uncover mechanisms.

Our Approach

We integrated diverse omics data, adjusted for confounders and uncovered mechanisms of supplementation with advanced analyses and visualisations.

• Multi‑omics integration & confounder adjustment: Collected and annotated omics datasets, assessed statistical significance with FDR correction and built a pipeline to remove confounder effects and correlate within and across omics.
• Mechanism discovery & visualisation: Identified biological mechanisms and microbial pathways through enrichment and causal mediation analyses, and produced publication‑quality visualisations using an expert team.

The Outcome

• Developed a flexible analytics pipeline applicable to multiple omic types.
• Extracted valuable information on mechanisms underpinning maternal and infant health.
• Identified concerted microbial pathways and produced high‑quality visualisations.
• Work contributed to a series of publications in peer‑reviewed journals.

Why It Matters

Multi‑omics integration can reveal complex biological mechanisms that single‑omic analyses miss. This project provides a reusable framework for nutrition and health studies and demonstrates the power of combining machine learning, statistics and domain expertise.