Metabolism Regulation Atlas: linking signalling and metabolism through post-translational regulation of metabolic enzymes

Abstract Background/Motivation: Advances in comprehensive reconstructions of human metabolism and signalling pathways provide a basis for systematic analysis of mechanisms connecting cell signalling to metabolic reactions. At the same time, information about these connections is scattered across different publications and pathway resources, limiting integrated mechanistic interpretation. Results Metabolism Regulation Atlas (MRegAtlas, https://metabolismregulation.github.io) was developed as a mechanistic bridge between signalling and metabolism, representing the regulatory context around metabolic enzymes. The maps in MRegAtlas link upstream signalling events, particularly post-translational modifications, to downstream metabolic reactions. The resource contains manually curated maps including regulation of glycolysis, glycogen metabolism, mevalonate metabolism, cholesterol metabolism, eicosanoid metabolism, nitric oxide production and phosphoinositide metabolism. The 21 maps comprise 230 processes and 546 nodes, including 190 proteins and 227 metabolites. Maps are represented in SBGN Process Description and are available for interactive exploration online via the MINERVA Platform. Methods The initial set of 17 maps in MRegAtlas were built from candidate connections identified by manual literature search. By combining Reactome Neo4j and Recon2 in Neo4j, we then implemented a method to systematically identify connections between metabolic and signalling networks with the focus on the regulation of metabolic enzymes. The identified connections were compared with the 17 maps, and four new maps were developed based on the candidates identified by the Neo4j framework: pyruvate dehydrogenase, branched-chain amino acid catabolism, urea cycle and nNOS maps. Conclusion MRegAtlas provides manually curated maps that link signalling events to metabolic enzymes. The maps can be used for visual analysis, education and computational model development, and the Neo4j framework is designed for continued mapping of the signalling mechanisms that regulate metabolic reactions.