Virtual screening is a proven method for drug-lead discovery and recent advances in the field have shown that upwards of 100M molecules can be screened to a target. However, for the average user, and as databases such as ZINC continue to grow, screening such large numbers of compounds is still quite challenging. As an alternative orthogonal approach, molecules can be assembled directly in a protein binding pocket using principles of de novo design. In this work, we evaluate two new de novo design methods recently implemented by our laboratory into DOCK6.9 which enable: (1) from scratch ligand assembly using a library of sidechains, linkers, and scaffolds (de novo growth) and (2) ligand refinement starting from an existing docked scaffold at user-defined points of modifications (de novo refinement). Results targeting fatty acid binding protein 5 (FABP5) and neutral sphingomyelinase 2 (nSMase2) will be presented with the goal of building ensembles of potential inhibitors for which subsets (or close analogs) may be commercially available for experimental testing. For both systems, structurally specific molecules were successful constructed using de novo DOCK that show favorable interactions with key binding site residues.

Fragment-based de novo design targeting FABP5 and nSMase2

2019 Fall National Meeting

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