Restricted cooperative games on metabolic networks reveal functionally important reactions

Understanding the emerging properties of complex biological systems is in the crux of systems biology studies. Computational methods for elucidating the role of each component in the synergetic interplay can be used to identify targets for genetic and metabolic engineering. In particular, we aim at determining the importance of reactions in a metabolic network with respect to a specific biological function. Therefore, we propose a novel game-theoretic framework which integrates restricted cooperative games with the outcome of flux balance analysis. We define productivity games on metabolic networks and present an analysis of their unrestricted and restricted variants based on the game-theoretic solution concept of the Shapley value. Correspondingly, this concept provides a characterization of the robustness and functional centrality for each enzyme involved in a given metabolic network. Furthermore, the comparison of two different environments – feast and famine – demonstrates the dependence of the results on the imposed flux capacities.

► Cooperative game theory and FBA are combined to investigate metabolic networks.
► The Shapley value is utilized as a relevance measure of individual reactions.
► Unrestricted and restricted game formulations are investigated.
► The Shapley value quantifies robustness and functional centrality.
► A model of monosaccharide metabolism of E. coli is analyzed.