Learning representations of microbe-metabolite interactions

Nat Methods. 2019 Dec;16(12):1306-1314. doi: 10.1038/s41592-019-0616-3. Epub 2019 Nov 4.


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.

Publication types

  • Research Support, N.I.H., Extramural
  • Research Support, Non-U.S. Gov't
  • Research Support, U.S. Gov't, Non-P.H.S.

MeSH terms

  • Animals
  • Bacteria / metabolism*
  • Benchmarking
  • Cyanobacteria / metabolism
  • Cystic Fibrosis / microbiology
  • Inflammatory Bowel Diseases / microbiology
  • Mice
  • Microbiota*
  • Neural Networks, Computer
  • Pseudomonas aeruginosa / metabolism

Supplementary concepts

  • Microcoleus vaginatus