By the time of this meeting, we expect ZINC will contain more than 2.5 billion commercially available, rule-of-4 (Ro4), biologically relevant compounds that can be purchased and tested within two months. With the help of commercial compound suppliers, we expect purchasable chemical space to continue to grow sharply for the foreseeable future. By summer 2020, we plan to have over 1 billion Ro4 3D molecules ready for large scale docking. We are focusing on two approaches to access this new and growing chemical space. First, molecular docking is a pragmatic approach to use protein structure to discover new small molecules to modulate protein activity. Successful projects against over 100 targets have now been reported in the literature. Our recent work on AmpC beta-lactamase and Dopamine D4 (Lyu, Nature, 2019) showed that docking screens of large libraries can find novel chemotypes and compounds with high affinity. A second approach to ligand discovery is analog-by-catalog. On average, compounds in ZINC have 20 close analogs, often allowing for rapid exploration of SAR around many hits when they are found. A key problem is being able to search the database in real time as the database grows. We have recently begun using Arthor and Smallworld by Nextmove Software to provide rapid public searching capabilities for ZINC. Together, we use these two approaches to visualize and prioritize the parts of purchasable chemical space that are of interest for our target. The scalability of our approaches to the planned future growth of purchasable chemical space will be discussed.

Using DOCK and ZINC to visualize ultra-large chemical libraries

2019 Fall National Meeting

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.