The recently developed AlphaFold 2 (AF2) method predicts protein's 3D structures from amino acid sequences. The open AF2 structure database covers the complete human proteome. It shows great potential to provide structural information to enable and enhance existing and new drug discovery projects. Using an industry-leading molecular docking method, Glide, we benchmark the virtual screening performance of 29 common drug targets between AF2 structures and known holo/apo structures from PDB. The AF2 structures show comparable early hit enrichments (avg. EF 1%: 12.74) with apo structures (avg. EF 1%: 11.16), while falling behind the enrichment of the holo structures (avg. EF 1%: 24.65). We also demonstrated that with an induced-fit docking approach, we can refine the AF2 structures to improve the performance in structural based virtual screening (avg. EF 1%: 18.89). Thus, with proper preparation and refinement, AF2 structures can be used to enable structural based hit identifications, expanding the number of prosecutable proteins.

Evaluating the use of AlphaFold 2 protein structures to expand available targets amenable to structure-based virtual screening

Fall 2022 abstracts, posters, and presentations will launch in mid-September.

ACS Fall 2022

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