(648a) Community Data-Driven Metabolic Network Reconstruction for Predicting Interspecies Interactions in a Photoautotroph-Heterotroph Consortium | AIChE

(648a) Community Data-Driven Metabolic Network Reconstruction for Predicting Interspecies Interactions in a Photoautotroph-Heterotroph Consortium

Authors 

Song, H. S. - Presenter, Pacific Northwest National Laboratory
Bernstein, H. C., Pacific Northwest National Laboratory
Weisenhorn, P., Argonne National Laboratory
Taylor, R. C., Pacific Northwest National Laboratory
Henry, C. S., Argonne National Laboratory
Zucker, J., Pacific Northwest National Laboratory
Microbial communities exert a significant impact on our environment, human health and industry. A mechanistic understanding of complex interplay between member species is essential for engineering microbial communities. Metabolic network models that provide comprehensive community-wide predictions on cross-species metabolite exchanges can serve as a useful tool for that purpose. Compared to single species analyses, community metabolic network reconstruction is more complex because it has to account for interspecies interactions. Current approaches focus on reconstruction of high-quality individual networks so that combined networks can subsequently lead to the prediction of interspecies interactions and community behaviors. This method becomes ineffective, however, for environmental communities whose members cannot be sufficiently characterized due to the difficulty in isolation and axenic cultivation. In order to address this limitation, we here tested a new approach that uses community-level data as a critical input for network reconstruction. Incorporation of community data is advantageous because it provides information on interspecies metabolic interactions, which is not necessarily obtainable from axenic cultures. To evaluate the validity of the proposed method, we reconstructed community networks based on alternative strategies of gapfilling (i.e., individual vs. community-level gapfilling). We implemented all processes for network reconstruction and refinement using the DOE Systems Biology Knowledgebase (KBase) platform (www.kbase.us). In the case study of a photoautotroph-heterotroph binary consortium, metabolic networks that were gapfilled and refined using community data provided experimentally validated predictions of how a photoautotrophic cyanobacterium provides organic carbon and nitrogen sources to support the growth of an obligate heterotrophic species. Further developments are in progress to extend the proposed method to more complex communities beyond binary consortia.