High-Throughput Virtual Screening of Small Molecule Inhibitors for SARS-CoV-2 Protein Targets with Deep Fusion Models

Stevenson, Garrett A.; Jones, Derek; Kim, Hyojin; Bennett, W. F. Drew; Bennion, Brian J.; Borucki, Monica; Bourguet, Feliza; Epstein, Aidan; Franco, Magdalena; Harmon, Brooke; He, Stewart; Katz, Max P.; Kirshner, Daniel; Lao, Victoria; Lau, Edmond Y.; Lo, Jacky; McLoughlin, Kevin; Mosesso, Richard; Murugesh, Deepa K.; Negrete, Oscar A.; Saada, Edwin A.; Segelke, Brent; Stefan, Maxwell; Torres, Marisa W.; Weilhammer, Dina; Wong, Sergio; Yang, Yue; Zemla, Adam; Zhang, Xiaohua; Zhu, Fangqiang; Lightstone, Felice C.; Allen, Jonathan E.

Structure-based Deep Fusion models were recently shown to outperform several physics- and machine learning-based protein-ligand binding affinity prediction methods. As part of a multi-institutional COVID-19 pandemic response, over 500 million small molecules were computationally screened against four protein structures from the novel coronavirus (SARS-CoV-2), which causes COVID-19. Three enhancements to Deep Fusion were made in order to evaluate more than 5 billion docked poses on SARS-CoV-2 protein targets. First, the Deep Fusion concept was refined by formulating the architecture as one, coherently backpropagated model (Coherent Fusion) to improve binding-affinity prediction accuracy. Secondly, the model was trained using a distributed, genetic hyper-parameter optimization. Finally, a scalable, high-throughput screening capability was developed to maximize the number of ligands evaluated and expedite the path to experimental evaluation. In this work, we present both the methods developed for machine learning-based high-throughput screening and results from using our computational pipeline to find SARS-CoV-2 inhibitors.