PSC COVID-19 Update: April 13, 2020

University of Virginia Simulating COVID-19 Interventions on Bridges

Researchers at the Biocomplexity Institute, University of Virginia, are working closely with PSC to perform highly detailed epidemiological studies of COVID-19 in support of U.S. federal agencies’ response and resource allocation efforts. For their simulations, more than 720 high-performance computing (HPC) servers on PSC’s Bridges supercomputing platform are dedicated nightly, using many nodes together to model populous states with complex social structures such as New York, California and Texas. The expected outcomes of the studies are nightly analyses of various counter-factual studies (studies testing interventions versus no interventions) of the spread of COVID-19, healthcare resource requirements and effectiveness of social distancing and other policies for “flattening the curve” for each county in the US.

“We thank NSF’s XSEDE project and PSC for their incredible help by making critical HPC resources promptly available to address the COVID-19 national emergency,” said Madhav Marathe, a division director and professor at the Biocomplexity Institute, who is coordinating the institute’s effort. “The NSF resources are critical for our work and enable science-based decision making. The excellent PSC staff were invaluable in making it possible to update our current simulations with the latest data and run them reliably and efficiently every night to be able to give decision makers the best possible information daily.”

PSC is providing Bridges and Bridges-AI to the national research community to perform urgent simulations aimed at reducing the spread of COVID-19 throughout the United States. 40% of Bridges is now dedicated to simulations of national importance. PSC's systems and user support staff are actively collaborating to rapidly execute these very large simulations. Bridges and the PSC staff contributing to the project are supported by the National Science Foundation.

COVID-19 research on Bridges.

COVID-19 HPC Consortium Projects Running on Bridges

PSC’s involvement with the national COVID-19 HPC Consortium continues, with several projects now running on the Bridges/Bridges-AI platform. The projects are applying artificial intelligence to search for new treatments for COVID-19 infections and performing detailed genomic analysis of the virus that causes COVID-19.

The consortium projects are being selected from among proposals made by the scientific community via an accelerated review process. Committees of experts in different scientific fields relevant to the pandemic meet daily at 11 a.m., submitting their recommendations to a “matching committee” that determines allocations to each project by that afternoon, with the computer available as early as the next morning.

Obtaining computing time on Bridges through the COVID-19 HPC Consortium.

Comprehensive COVID-19 Multi-omics Data Resource Available to Researchers

Offering Bridges as a one-stop Big Data portal for advanced computation applied to studies on the virus that causes COVID-19, PSC continues to make the 2019 Novel Coronavirus Resource (2019nCoVR) available for open research. This ever-growing data resource features comprehensive integration of worldwide genomic and proteomic sequences, as well as their metadata, from the Global Initiative on Sharing All Influenza Data (GISAID), National Center for Biotechnology Information (GenBank), China National GeneBank (CNGBdb), National Microbiology Data Center and China National Center for Bioinformation (CNCB)/National Genomics Data Center (NGDC). PSC will update the local resource twice a week, keeping the latest data on the virus available for research on virus classification and origin, genome variation and evolution, fast detection, drug development and pneumonia precision prevention and therapy to combat COVID-19.

For more information on gaining access to this valuable resource on Bridges, please visit PSC’s community datasets page.

Events Calendar

<<  <  July 2020  >  >>
 Su  Mo  Tu  We  Th  Fr  Sa 
     1  2  3  4
  5  6  7  8  91011