Metabolic Network Percolation Provides Insight into Microbial Fastidiousness

We present a new constraints based modeling method designed to provide quantitative insight into the dependence of microbial growth on the availability of specific environmental nutrients, also known as fastidiousness. Our method, which is inspired by percolation theory, uses a Monte Carlo approach to quantify the average environmental complexity necessary for a metabolic network to synthesize a specified metabolite (e.g. an amino acid) or set of metabolites (e.g. biomass). By performing incremental additions of random metabolites to the network, the algorithm calculates the critical threshold at which the network is capable of synthesis. Alongside our algorithm, we devised a combinatorics-based theory to analytically calculate this critical threshold as a function of the average complexity and redundancy of the pathways necessary to produce the product. This analytical work serves to clarify the mathematical underpinnings of our algorithm and its results. We tested the capacity of our method to capture the different biosynthetic capabilities of Escherichia coli and Mycoplasma genitalium, two organisms with drastically different reliance on exogenously supplied metabolites, and highlight our method’s ability to recapitulate meaningful biological metrics associated with amino acid biosynthesis. We next adapted our method towards analyzing automatically-generated, raw (i.e. non-gap-filled) stoichiometric models from the DOE Systems Biology Knowledgebase (KBase.us). Our method can thus provide biological insights that do not depend on specific assumptions of environment composition and gap-filling criteria, and is a promising tool for the study of partially curated metabolic models such as those built from metagenomes of complex communities. Moving forward, our method can be further extended to the study of interactions between species and of community-level fastidiousness, which will provide valuable insights into the nature of complex microbial communities.