UC Berkeley Cloud Meetup 031: COVID-19 research using data science & Azure

January 27, 2022

Details:

Using data science and the cloud to better understand coronaviruses and COVID-19

This month's meetup showcases two UC Berkeley research groups’ use of Azure cloud services to meet research challenges of their award as part of the C3.ai Digital Transformation Institute’s Mitigation of COVID-19 and Future Pandemics call for proposals. In one case a researcher “took ownership” of the group’s cloud deployment, and in another case Institute staff deployed cloud infrastructure behind “one-button access” for researchers.

“One-Button Access” to Cloud-hosted Research on COVID-19

Yana Yarmolich (UC Berkeley undergraduate student) will present research and team-management work under the supervision of Dean Emeritus and Professor Stefano Bertozzi analyzing COVID-19 data from Mexico. Yana will demonstrate use of the group’s Azure cloud deployment that the group uses to research the transmission of SARS-COV-2 and the clinical presentation of COVID-19 in Mexico.

DTI DevOps Lead Greg Merritt will present a high-level look “under the hood” of the deployed cloud infrastructure.

Using Data Science and Cloud Computing for Proposing SARS-COV-2 Inhibiting Drugs

Jerry Li (UC Berkeley graduate student) will present research and cloud deployment work under the supervision of Professor Teresa Head-Gordon, Chancellor's Professor, Departments of Chemistry, Bioengineering, Chemical & Biomolecular Engineering. Jerry will describe the group’s research problem (“Reinforcement Learning with Real-time Docking of 3D Structures to Cover Chemical Space: Mining for Potent SARS-CoV-2 Main Protease Inhibitors”) and the efficient use of multiple Azure services (Azure ML, Azure Functional App and Azure Batch) to carry out CPU-GPU heterogeneous computations, specifically using Azure Batch to perform massive-parallelized CPU calculations.