The Convergence of HPC, AI and Big Data in Rapid-Response to the COVID-19 Pandemic

The "Force for Good" pledge of intellectual property to fight COVID-19 brought into action HPE products, resources and expertise to the problem of drug/vaccine discovery. Several scientists and technologists collaborated to accelerate efforts towards a cure. This paper documents the spirit of such a collaboration, the stellar outcomes and the technological lessons learned from the true convergence of high-performance computing (HPC), artificial intelligence (AI) and data science to fight a pandemic. The paper presents technologies that assisted in an end-to-end edge-to-supercomputer pipeline - creating 3D structures of the virus from CryoEM microscopes, filtering through large cheminformatics databases of drug molecules, using artificial intelligence and molecular docking simulations to identify drug candidates that may bind with the 3D structures of the virus, validating the binding activity using in-silico high-fidelity multi-body physics simulations, combing through millions of literature-based facts and assay data to connect-the-dots of evidence to explain or dispute the in-silico predictions. These contributions accelerated scientific discovery by: (i) identifying novel drug molecules that could reduce COVID-19 virality in the human body, (ii) screening drug molecule databases to design wet lab experiments faster and better, (iii) hypothesizing the cross-immunity of Tetanus vaccines based on comparisons of COVID-19 and publicly available protein sequences, and (iv) prioritizing drug compounds that could be repurposed for COVID-19 treatment. We present case studies around each of the aforementioned outcomes and posit an accelerated future of drug discovery in an augmented and converged workflow of data science, high-performance computing and artificial intelligence.