Chemogenomic sets: valuable tools for early-stage drug discovery

Small-molecule chemical probes are foundational to biomedical research as they enable interrogation of the function of individual proteins within biological systems. However, the utility of these tools is constrained by their limited coverage of the proteome. To overcome this limitation, chemogenomic sets, collections of molecules annotated for their activity against a range of biological targets, have emerged to expand the coverage and interrogation of the proteome by small molecules. Unlike traditional chemical probes, these sets leverage annotated polypharmacology to allow for identification of therapeutic vulnerabilities in phenotypic screening. The present review summarizes the current landscape of chemogenomic sets, including open-science initiatives and target-class-focused libraries, and proteome-wide collections developed by academic and industrial efforts. We highlight key applications of these sets in phenotypic screening, target identification, and computational modeling, demonstrating how these resources allow for target identification in disease-relevant pathways. We envision that the data generated by these sets will facilitate the use of machine learning to fuel early-stage drug discovery.