An algorithm for designing minimal microbial communities with desired metabolic capacities

Alexander Eng, Elhanan Borenstein*

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review


Motivation: Recent efforts to manipulate various microbial communities, such as fecal microbiota transplant and bioreactor systems' optimization, suggest a promising route for microbial community engineering with numerous medical, environmental and industrial applications. However, such applications are currently restricted in scale and often rely on mimicking or enhancing natural communities, calling for the development of tools for designing synthetic communities with specific, tailored, desired metabolic capacities. Results: Here, we present a first step toward this goal, introducing a novel algorithm for identifying minimal sets of microbial species that collectively provide the enzymatic capacity required to synthesize a set of desired target product metabolites from a predefined set of available substrates. Our method integrates a graph theoretic representation of network flow with the set cover problem in an integer linear programming (ILP) framework to simultaneously identify possible metabolic paths from substrates to products while minimizing the number of species required to catalyze these metabolic reactions. We apply our algorithm to successfully identify minimal communities both in a set of simple toy problems and in more complex, realistic settings, and to investigate metabolic capacities in the gut microbiome. Our framework adds to the growing toolset for supporting informed microbial community engineering and for ultimately realizing the full potential of such engineering efforts.

Original languageEnglish
Pages (from-to)2008-2016
Number of pages9
Issue number13
StatePublished - 1 Jul 2016
Externally publishedYes


FundersFunder number
National Center for Complementary and Integrative HealthDP2AT007802


    Dive into the research topics of 'An algorithm for designing minimal microbial communities with desired metabolic capacities'. Together they form a unique fingerprint.

    Cite this