Atom-Resolved Genome-Scale Metabolic Models of the Human Gut Microbiome

Abstract Detailed genome-scale metabolic models (GEMs) are essential for math-ematical simulation of metabolism, with applications in industry andresearch, including bacterial community models, host-disease interactionmodels, and numerous medical applications.Here, we provide GEMs of microbial species from 7 phyla of the humangut microbiome, with novel, detailed annotations of metabolic reactionsusing validated molecular structures and complete atom-to-atom map-pings (AAMs).By incorporating structural information with AAMs, GEM applicationsgo beyond traditional stoichiometric analyses and enable atomic-level ex-amination of biochemical transformations, such as isotope-labeling stud-ies, metabolic flux tracing, and investigation of metabolic interactionsin gut microbial communities. The dataset additionally contains 65metabolic pathways, each with complete AAMs, enabling the indepen-dent modeling and analysis of individual aspects of microbial metabolism.This dataset provides a basis for developing, validating, and applyingfull-resolution metabolic models by combining genome-scale metabolicmodeling with detailed information about atomic-level reactions.