Thanks to collaborations with commercial and academic research partners alike, the Argonne National Laboratory has published new research showing the capabilities of artificial intelligence and simulation workflows to provide insight to the SARS-CoV-2 viral genome and potentially fuel drug development.
SARS-CoV-2, the virus that causes COVID-19, includes two proteins — nsp10 and nsp16 — within its viral genome, which help the virus to replicate and evade immune responses. Collaborators utilized two hardware platforms — Cerebras CS-1 and ThetaGPU — to investigate the pair, with assistance from a newly developed application of an AI method known as deep learning. They then used this method to push adaptive molecular dynamics simulations and stream these simulations from ThetaGPU onto the Cerebras platform to simultaneously analyze protein interactions.
“This needs to be done at a scale that is unprecedented since the data generation and AI components have to run side-by-side,” Arvind Ramanathan, Argonne computational biologist and a member of the research team, said. “The idea is, if one machine is good at doing MD simulations and another is very good at AI, then why not couple the two to produce a much larger system that offers more throughput with AI.”
Such actions could, according to Ramanathan, open the door to advancing knowledge of protein function at large. It could also enable drug discoveries to treat SARS-CoV-2 and other diseases, thanks to greater characterization of the physical processes providing the foundations of specific biological functions. As more complex models are developed, the research could even aid vaccine development.
“In the context of the SARS-CoV-2 virus, a fundamental understanding of molecular processes, such as the nsp16-nsp10 interaction, is important if we want to design drugs that can stop the virus in its path,” Ramanathan said.
The team’s research was published in the proceedings from the Platform for Advanced Scientific Computing Conference in July. Research was undertaken through collaboration between Argonne and Cerebras Systems Inc., with support from the U.S. Department of Energy-run Exascale Computing Project and the Virtual Biotechnology Lab. Funding was made possible with support from the federal CARES Act.