An evolutionary algorithm for designing microbial communities via environmental modification

Despite a growing understanding of how environmental composition affects microbial community properties, it remains difficult to apply this knowledge to the rational design of synthetic multispecies consortia. This is because natural microbial communities can harbor thousands of different organisms and environmental substrates, making up a vast combinatorial space that precludes exhaustive experimental testing and computational prediction. Here, we present a method based on the combination of machine learning and dynamic flux balance analysis (dFBA) that selects optimal environmental compositions to produce target community phenotypes. In this framework, dFBA is used to model the growth of a community in candidate environments. A genetic algorithm is then used to evaluate the behavior of the community relative to a target phenotype, and subsequently adjust the environment to allow the organisms to more closely approach this target. We apply this iterative process to in silico communities of varying sizes, showing how it can rapidly identify environments that yield desired phenotypes. Moreover, this novel combination of approaches produces testable predictions for the in vivo assembly of microbial communities with specific properties, and can facilitate rational environmental design processes for complex microbiomes.