Genome-scale Metabolic Model Reconstruction of a Small Defined Soil Microbial Community

Abstract

Microbial communities play central roles in the functioning of ecosystems, engineering, and industry, while also contributing to the physiology and health of all living beings. Destabilized communities are prone to malfunctioning and invading pathogens. It is currently unknown, which processes or mechanisms cause communities to function properly or become dysfunctional. This lack of understanding hinders the efforts to reliably manipulate, restore function, or add novel metabolic capacities to microbial communities, especially in soil pollutant bioremediation and agriculture. One of the current trends to better understand the function of microbial community is to predict interspecific interactions and functional gains from genome-scale metabolic models. This thesis project aims to develop genome-scale metabolic models (GEMs) of 21 isolated soil bacteria isolates through genome assembly, annotation, and metabolic network inference. A single selected model for a soil Caulobacter was manually curated. Quality assessment of sequencing reads and assembled genomes showed good quality. Comparison of draft GEMs showed high functional redundancy of their core metabolisms, and more variability in ‘peripheral’ metabolic reactions, highlighting the unique niches of the isolates. Isolates further functionally grouped according to the taxonomic Class level, which might be a factor explaining soil community stability against disruptions by functional redundancies. The manually curated Caulobacter model cauj resulted in an estimated growth rate of 3.7141 mmol/(gDW·h), but needs further curation to remove spontaneous flux-bearing reactions in absence of carbon substrate. The current results are not yet sufficient to understand the specific interactions within the synthetic soil community. However, it gives a broad overview of the structural and functional capacity of the community.