Learning representations of microbe–metabolite interactions

Integrating multiomics datasets is critical for microbiome research; however, inferring interactions across omics datasets has multiple statistical challenges. We solve this problem by using neural networks ( https://github.com/biocore/mmvec ) to estimate the conditional probability that each molecule is present given the presence of a specific microorganism. We show with known environmental (desert soil biocrust wetting) and clinical (cystic fibrosis lung) examples, our ability to recover microbe–metabolite relationships, and demonstrate how the method can discover relationships between microbially produced metabolites and inflammatory bowel disease. mmvec, a neural-network-based algorithm, uses paired multiomics data (microbial sequence counts and metabolite abundances) to compute the conditional probability of observing a metabolite in the presence of a specific microorganism.