Profiling Biological Effects of Microbiome Metabolites via Machine Learning
Preprint
- 4 August 2025
- preprint
- Published by American Chemical Society (ACS) in ChemRxiv
Abstract
Human microbiome-derived metabolites are key mediators of host physiology. However, their biological effects remain largely uncharacterized due to limitations of current low-throughput and untargeted experimental approaches that are time-intensive and costly. This has hindered the systematic biological characterization of microbiome metabolites. To address this gap and accelerate the identification of biological effects of microbiome metabolites, we developed a machine learning platform trained on publicly available drug development data to rapidly predict a wide array of chemical and biological properties of microbiome metabolites. Prospective experimental validation confirmed the predictive accuracy of our models and uncovered previously unknown effects of several metabolites. For example, we identified previously unknown Interleukin-8 secretion stimulation by the metabolites spermine and spermidine, which have been regarded anti-inflammatory thus far. Our findings demonstrate the power of machine learning to accelerate the functional annotations of microbiome-derived metabolites, paving the way for new biomarker discovery and therapeutic development.Keywords
All Related Versions
- Version 1, 2025-08-04, ChemRxiv
- Published version: iScience, 29 (4), 115282.
This publication has 38 references indexed in Scilit:
- Gut microbiota-derived metabolites and their importance in neurological disordersMolecular Biology Reports, 2022
- The metabolic and toxic acute effects of phloretin in the rat liverChemico-Biological Interactions, 2022
- Deciphering the Chemical Lexicon of Host–Gut Microbiota InteractionsTrends in Pharmacological Sciences, 2019
- Treg Cells: A LAGging Hand Holds the Double-Edged Sword of the IL-23 AxisImmunity, 2018
- Topographic Reorganization of Cerebrovascular Mural Cells under Seizure ConditionsCell Reports, 2018
- Prediction of liver toxicity and mode of action using metabolomics in vitro in HepG2 cellsArchives of Toxicology, 2017
- Deep Learning for Drug-Induced Liver InjuryJournal of Chemical Information and Modeling, 2015
- Specialized Metabolites from the Microbiome in Health and DiseaseCell Metabolism, 2014
- Experimental and computational approaches to estimate solubility and permeability in drug discovery and development settings 1PII of original article: S0169-409X(96)00423-1. The article was originally published in Advanced Drug Delivery Reviews 23 (1997) 3–25. 1Advanced Drug Delivery Reviews, 2001
- Progress in predicting human ADME parameters in silicoJournal of Pharmacological and Toxicological Methods, 2000