To gain a better understanding of the relationship between feed parameters, biomethanation communities, and the employed metabolic pathways that lead this carbon reduction, data from a lab-scale biomethanation reactor was processed and integrated into a new pipeline that leverages metabolic modeling (through linear optimization augmented with ML) to elucidate the active pathways and ecological roles of each community member. This pipeline a) matches the 16S RNA sequences to their closest hits in a database of reference genomes; b) assemblies the union of metabolic functions associated with the reference genomes of all closest hits; c) reconstructs a genome-scale metabolic model (GEM) from each of these union of metabolic functions, which best captures the metabolism of the organism from which the 16S came; d) assembles these GEMs into a single compartmentalized community model (cGEM) with the experimentally observed member abundances; and e) simulates the cGEM via the Flux Balance Analysis algorithm over the experimental media.
The results from the above pipeline identify the critical members and pathways that reduce carbon from nothing more than 16S RNA sequences and media information, thereby opens the possibility for stably optimizing the purity and yield of biomethane from this system towards a circular bioeconomy. This pipeline has broad implications for automating the process for gleaning basic understanding of microbial communities from simple experimental measurements, and will be implemented into a KBase Application with point-and-click functionality that further streamlines this pipeline and accelerates bioinformatics research.