Targeting SARS-CoV-2 with AI- and HPC-enabled Lead Generation: A First Data Release

• DOI: 10.48550/arxiv.2006.02431
• Publication Date: 2020-01-01
• Authors: Yadu Babuji; Ben Blaiszik; Tom Brettin; Kyle Chard; Ryan Chard; Austin Clyde; Ian Foster; Zhi Hong; Shantenu Jha; Zhuozhao Li; Xuefeng Liu; Arvind Ramanathan; Yi Ren; Nicholaus Saint; Marcus Schwarting; Rick Stevens; Hubertus J. J. va...; Rick Wagner
• Keywords: Lead (geology); 2019-20 coronavirus outbreak

Researchers across the globe are seeking to rapidly repurpose existing drugs or discover new counter novel coronavirus disease (COVID-19) caused by severe acute respiratory syndrome 2 (SARS-CoV-2). One promising approach is train machine learning (ML) and artificial intelligence (AI) tools screen large numbers of small molecules. As a contribution that effort, we aggregating numerous molecules from variety sources, using high-performance computing (HPC) computer diverse properties those molecules, computed ML/AI models, then resulting models for screening. In this first data release, make available 23 datasets collected community sources representing over 4.2 B enriched with pre-computed: 1) molecular fingerprints aid similarity searches, 2) 2D images enable exploration application image-based deep methods, 3) 3D descriptors speed development models. This release encompasses structural information on 60 TB pre-computed data. Future releases will expand include more detailed simulations, other products.