Growing phylometabolic trees from metabolic networks

Comparative genomics has been essential to inferring the evolutionary trajectories of species and producing classifications on the basis of DNA sequence. However, these genomic classifications present an incomplete characterization of the functional role of species in an ecological environment. For example, changes to nutrient availabilities lead to changes in the metabolic fluxes of an organisms which requires a classification scheme that uses metabolic information. Previous work has shown the potential for classification methods using metabolic networks, but these do not directly utilize the full topological information contained in the networks. Here we introduce and apply a computational framework for a classification scheme of species that compares metabolic networks using distance metrics on linear subspaces of stoichiometric matrices. We find that for the AGORA2 human gut microbiome, the underlying genetic information is insufficient to match species to their metabolic classifications. We further demonstrate that (2+1)-dimensional phylometabolic trees constructed from the distances more faithfully capture relationships between species compared to standard planar representations.