Conventional protein-ligand docking algorithms rely on predefined targets in the protein to search for potential ligands with optimal binding. In this project, we propose to deploy an AI-driven computational approach on Summit to comprehensively identify and evaluate targets without pre-selection. In response to the outbreak of COVID-19, we apply our method to study the current human coronavirus (SARS-CoV-2) spike protein, the host protein to which it binds (ACE2 receptor), and the host enzyme used for priming (cellular serine protease TMPRSS2). By comparing these structures with the analogous machinery used by previous human coronaviruses (e.g., MERS-CoV, SARS-CoV), we will identify the specific regions of SARS-CoV-2 spike protein, ACE2 receptor, and TMPRSS2 which can potentially serve as drug targets. Moreover, the AI-driven pipeline developed in this work will be extensible to study other biomolecular systems including future viruses.
This project is part of the COVID-19 HPC Consortium.
|Number of Jobs Run To-Date||Node Hours Used To-Date||Number of Jobs Running Now||Nodes in Use Now|