Three bacteria genome, two from lactic bacteria and one from propionic bacterium. bacteria genome

Understanding the interactions within complex bacterial communities is a major issue in the modelling world. Such interactions occur mainly at metabolic level and computational models enable us to identify them. Two major steps are needed: building a genome-scale metabolic network (GEM) for every member of the community, then, reveal interactions between them. A metabolic network is a set of chemical reactions inferred from a genome and used to characterize the phenotype. In this study, GEMs of three bacteria of a cheese model are built using a top-down approach with carveMe tool. They form a resource for the identification of key metabolites and production pathways of organoleptic compounds, as well as regulation elements that we detect with mathematical formalisms. In addition, we constrained GEMs with multi-omics data such as metatranscriptomics, providing more in-depth predictions on the organization of the community. Models built are in accordance with the literature and show metabolite dependances within the bacterial community. Some dynamical simulations of the three species metabolism helps us to follow interactions through time and target key functions. Finally, constrain GEM and scaling up the mathematical models to more complex bacterial communities with the hybrid method remains to be done.