Identifying Novel Bioactive Microbial Gene Products in Inflammatory Bowel Disease

The gut microbiome and associated bioactive compounds are often disrupted in gastrointestinal conditions such as the inflammatory bowel diseases (IBD). Even in well-characterized environments (e.g. the human gastrointestinal tract), more than one-third of microbial proteins are uncharacterized. We prioritized potentially bioactive proteins from the human gut microbiome during IBD, beginning with a catalog of 1.6M protein families assembled from 1,595 metagenomes in the Integrative Human Microbiome Project (HMP2). ~70% of these proteins were uncharacterized, including those with strong homology to known functionally uncharacterized proteins (24%), new proteins with weak homology (33%), and completely novel proteins without homology (12%). We assigned putative annotations to uncharacterized proteins using a combination of guilt-by-association, taxonomic binning, secondary structure analysis, and host phenotype analysis, ultimately leaving only 30,529 families (3%) still functionally and taxonomically uncharacterized. >340,000 protein families were specifically prioritized as potentially bioactive with respect to gut inflammation by integrating evidence from host disease phenotypes, ecological properties. Strikingly, ~23% of them were novel proteins, 36% of which expanded the pangenomes of common gut taxa and 90% of the remainder were assigned at least one putative biochemical annotation. Our analysis methods are generalizable to other microbial communities and human disease, and we provide an open source implementation as MetaWIBELE. The prioritized results suggest thousands of new candidate microbial proteins likely to interact with host immunity in IBD and gut inflammation, offering a rational and targeted compendium of potential therapeutic compounds.