A University of Alabama in Huntsville (UAH) professor and his team are collaborating with the Oak Ridge National Laboratory (ORNL) supercomputer research initiative capitalizing on advanced, high-powered computing power to screen compounds for effectiveness against the COVID-19 virus. Time is of the essence as a pandemic spreads throughout America and Dr. Jerome Baudry is applying SUMMIT, the world’s most powerful supercomputer, to examine compounds that may have COVID-19 efficacy.
What is SUMMIT?
The SUMMIT supercomputer at ORNL was developed by IBM and by November 2019, as the world’s fastest supercomputer capable of 200 petaFLOPS. Its current LINPACK benchmark is clocked at 148.6 petaFLOPS. It is reported to be the third most energy efficient in the world. It is the world’s first supercomputer to reach exaop (a quintillion operations per second) speed, achieving 1.88 exaops during a genomic analysis and is expected to reach 3.3 exaops using mixed precision calculations.
Although still in the early stages of this research endeavor, the pressure is on and the stakes are big. Solutions to the COVID-19 problem must be solved. Hence the teams, Dr. Baudry reports, will perform “millions of virtual screening calculations per day.” It is an ‘all hands-on deck’ effort also including UAH supercomputers as well.
Living in a pandemic change one’s sense of urgency as COVID-19 ripples through America and the specter of a spike in deaths become more imminent—the nation’s leading universities, government laboratories and government agencies collaborate in myriad ways to help discover new approaches to stop this dangerous virus. Now on a daily basis the UAH professor, colleagues and ORNL team members get on teleconferences on what he describes as a “24/7” effort. The Northern Alabama biotech hub buzzes around the clock as the players know the stakes are big.
Focus: Repurposing Drugs
Taking a pragmatic approach, the UAH and ORNL teams are leveraging SUMMIT to repurpose existing drugs—“to take existing drugs from the shelf and find which ones are active against either the virus itself or can help in treating or mitigating the effects of infection in severe cases.” Hence, the team is either using SUMMIT on available drugs with safe profiles or natural products and also plan on studying compounds identified as future drugs. In this initial stage the team will not develop new drugs.
According to Baudry, they use the supercomputer to crank through the entire proteome of the virus to assess what the virus’ genome actually produces and essentially rapidly build computational models mirroring the virus’ protein output. He emphasized that they will continuously “repeat the repurposing process for each of these proteins.” They start with select proteins on the virus’ surface in an effort to stop it from infecting human cells. Apparently, certain proteins actually enable the virus to copy itself once it penetrates the human cell in an endeavor to actually oppose this course in a comparison that Baudry likens to how anti-AIDS drugs work. However by plan on using SUMMIT to expand to as much of the virus’ genome as possible.
UAH Lab All-in
Dr. Baudry also includes a high-performance computer cluster built in collaboration with UAH’s Office of Information Technology. They plan on executing thousands of virtual screenings per day in conjunction with the SUMMIT collaboration. Apparently, the team here in Alabama is known for their expertise in this field including a Dr. Kendall Byler a seasoned expert in matters of using these supercomputers to analyze and screen for prospects.
The Alabama Compute/Biotech Cluster
North Alabama has a strong concentrated capability combining supercomputing and biotech assets, reports Baudry. In addition to UAH and its internal capability (and its collaboration with Oak Ridge just to the north), they will leverage the Alabama Supercomputer Center in Huntsville “to ‘level up our computations as well.” Baudry noted additionally, “If we could run the Alabama Supercomputing Center, we could perform several hundred thousand virtual screenings a day.” Hence, the goal would be to leverage ORNL for the biggest computing jobs and parsing out sub-calculations to UAH and Alabama high-performance supercomputers.
Dr. Jerome Baudry, the Ms. Pei-Ling Chan Chair in the Department of Biological Sciences, University of Alabama, Huntsville