Skip to content

The network

Origin

NUTRIx grew out of a spring school on machine learning in nutrition, funded by the Joachim Herz Foundation and organised by Mattea Müller and Jan Voges in 2024. Many of the network's members first met there and recognised a shared set of methodological challenges — and opportunities for collaboration — at the intersection of nutrition and data science.

Rationale

Nutrition science is inherently interdisciplinary and is undergoing a methodological transformation driven by digitalisation and high-dimensional data. Modern datasets — digitised dietary intake records, continuous glucose and activity monitoring, metabolic and inflammatory biomarkers, and multi-omics layers (genomics, transcriptomics, proteomics, metabolomics, microbiome) — open the way to a mechanistic, systems-level understanding of how diet influences health.

Focus areas

The network concentrates on three interrelated areas:

  1. Machine learning applied to dietary intake and survival data — prediction versus causal inference, estimands, confounding and DAGs, substitution models, non-linear associations, and N-of-1 designs.
  2. Transparency and reproducibility in ML-driven nutrition research — best practices, data quality, avoiding data leakage, uncertainty and stability, documentation, and FAIR-aligned, shareable workflows.
  3. Integrative analysis of multi-omics datasets in a nutritional context — harmonisation, preprocessing, batch effects, dimensionality reduction, multi-omics integration algorithms, and network-based interpretation.

Objectives

  1. Strengthen interdisciplinary collaboration at the interface of nutrition and data science through regular, structured exchange between early-career researchers across institutions.
  2. Address core methodological challenges in computational nutrition science: dietary intake and survival data, multi-omics integration, and reproducibility of ML workflows.
  3. Promote open science by collaboratively building this public, open-source knowledge base of reproducible pipelines, tutorials, and methodological guidance.
  4. Support early-career development through peer learning, engagement with international experts, and a transatlantic collaboration with the NIH Precision Nutrition Training Programme.
  5. Contribute to the wider community through a symposium and a public online workshop, and through alignment with national and European infrastructures such as ELIXIR.