Partitioning of microbial function among taxonomic ranks across the tree of life

Concatenated protein sequence alignments are a reliable tool to determine phylogenetic relationships among microbes1 but do not necessarily imply microbial function. Microbiomes influence their environments via the collective action of microbial metabolisms, so it is important to develop a quantitative understanding of the degree to which taxonomy provides information about the function of microbes. However, the overwhelming majority of microbial lineages are resistant to culturing, leading to difficulties characterizing their metabolisms. Here we quantify the enrichment of clusters of orthologous gene functional categories (COG-FCs), a proxy for metabolic potential, within the genomes of 3207 cultured and uncultured microbe lineages from a custom-curated genome database. Results show that, on average, 36.7% of the variation in COG-FC enrichment is explained by taxonomic rank, with domain, phylum, class, order, family, and genus explaining, on average, 2.3%, 12.9%, 2.6%, 9.4%, 6.1%, and 3.5% of the variance, respectively (p<0.001 for all). Furthermore, taxonomic rank explained 81.5% of the variance when a only 18 of the 23 COG-FC categories were considered for analysis. To our knowledge, this is the first work which quantifies the variance in metabolic potential from individual taxonomic ranks. A qualitative comparison between the COG-FC enrichments and order-level phylogenies, generated from concatenated protein sequence alignments, further supports the idea that metabolic potential is taxonomically coherent at higher taxonomic ranks. The coherence in metabolic potential at higher taxonomic ranks provides support to hypotheses which suggest that microbial lineages may be biased to occupy specific ecological or metabolic niches.