ABSTRACT A complex interplay exists between the human microbiome and the host, resulting in clear effects on human physiology and microbiome ecology. A promising avenue to dissect this interplay at a mechanistic level is through the study of microbiome-derived molecules that mediate important microbe-microbe and microbe-host interactions. In this application, we propose a hybrid computational- synthetic biology approach to discover, rationally prioritize and systematically characterize microbiome- derived molecules. We propose to apply this approach to three structurally diverse classes of bioactive molecules that are widely encoded by the human microbiome but remain severely understudied in terms of both structure and function. First, guided by the computational analysis of biosynthetic gene clusters in metagenomic sequencing data from the human microbiome of thousands of subjects, we will select specific members of the three molecular classes for experimental characterization. Second, we will use genome editing of native members of the microbiome and synthetic biology in a multi-host heterologous expression platform to characterize the selected pathways and their products. Finally, we will employ an array of in vitro, cell-based and mouse colonization assays to interrogate the role of the discovered molecules in mediating relevant host and microbiome functions. Taken together, our approach will unveil an undermined section of the interplay between the human microbiome and the host and provide diverse microbiome-derived bioactive molecules as tools for future mechanistic studies and therapeutic interventions.

NARRATIVE Human microbiome-derived bioactive molecules play an important role in mediating microbiome functions and host effects, but we have thus far only discovered a small number of these molecules. Here, we propose to systematically characterize the structure and function of three classes of diverse molecules that are widely encoded by the human microbiome but are currently understudied. Our findings will move us closer towards accurate prediction of microbiome function in health and disease, and allow us to identify potential entry points for targeted interventions.