Bayesian inference captures metabolite-bacteria interactions in a microbial community.

Macro-ecosystems, including the human gut, host a vast and diverse set of microbes that indirectly interact with each other through consuming and producing metabolites. Disruptions in this microbial network can affect macro-ecosystem functioning and, in the human gut, contribute to the onset and progression of various disorders, including diabetes, rheumatoid arthritis, and Parkinson's disease. A theoretical foundation for understanding the intricate and dynamic interactions between microbes and metabolites is essential for developing microbiota-targeted interventions to improve macro-ecosystem functioning and health. To this end, a precise mathematical framework is crucial to capture and quantify the complex dynamics of the microbial system. Here, we develop a dynamic network model of coupled ordinary differential equations and present a computational workflow that integrates a generative model with Bayesian inference for model identification. Our approach infers interaction rates, quantifying metabolite consumption and production from simulated time-series data within a Bayesian framework, incorporating prior knowledge and uncertainty quantification. We show that our approach is accurate and reliable in communities of various sizes, sparsity, and with different levels of observational noise. This workflow enables