Predicting Nash equilibria for microbial metabolic interactions

• DOI: 10.1093/bioinformatics/btaa1014
• Publication Date: 2020-11-25T02:43:35Z
• Authors: Jingyi Cai; Tianwei Tan; Siu H J Chan
• Keywords: Flux Balance Analysis; Mutualism

Microbial metabolic interactions impact ecosystems, human health and biotechnology profoundly. However, their determination remains elusive, invoking an urgent need for predictive models seamlessly integrating metabolism with evolutionary principles that shape community interactions.Inspired by the game theory, we formulated a bi-level optimization framework termed NECom which any feasible solutions are Nash equilibria of microbial with/without outer-level (community) objective function. Distinct from discrete matrix games, continuous interdependent strategy space fluxes. We showed successfully predicted several classical games in context were falsely or incompletely existing methods, including prisoner's dilemma, snowdrift cooperation. The improved capability originates novel formulation to prevent 'forced altruism' hidden previous static algorithms while allowing sensing all potential metabolite exchanges determine evolutionarily favorable between members, feature missing dynamic methods. results provided insights into why mutualism is despite seemingly costly cross-feeding metabolites demonstrated similarities differences flux games. was then applied reported algae-yeast co-culture system shares typical features lichen, model mutualism. 488 growth conditions corresponding 3221 experimental data points simulated. Without training parameters using data, more species' rates given uptake compared balance analysis overall 63.5% 81.7% reduction root-mean-square error two species respectively.Simulation code available at https://github.com/Jingyi-Cai/NECom.git.Supplementary Bioinformatics online.