New NIH Antiviral Center Expands Patent-pending ML Research

June 02, 2022

C3.ai DTI researcher Teresa Head-Gordon will continue the machine learning COVID-19 research spurred by a 2020 DTI grant, with the Midwest Antiviral Drug Discovery Center for Pathogens of Pandemic Concern, one of nine new centers announced by the National Institute of Allergy and Infectious Diseases on May 18, 2022.

“We are thrilled to be part of the Midwest AviDD center,” says Head-Gordon, professor of Chemistry, Bioengineering, and Chemical & Biomolecular Engineering at UC Berkeley. “My group has a large number of new ideas that combine physics and machine learning to develop new drug leads.”

Head-Gordon’s C3DTI project, “Scoring Drugs: Small Molecule Drug Discovery for COVID-19 using Physics-Inspired Machine Learning,” led to a provisional patent for novel reinforcement learning techniques that identified 54 molecules as potent Mpro inhibitors (7 with better synthetic potential), covering a much broader range than crowd-sourced projects such as the COVID moonshot.

“Our model could be quickly applied to design novel inhibitors for proteins relevant to other global diseases,” the team writes in a 2021 paper, citing as one example multiple targets for bacterial resistance to antibiotics. Concluding with, “Overall, we believe our tool will be of great benefit to the computational and medicinal chemistry fields at large, and potentially aid traditional drug-design workflows as well.”

Read more:
NIAID Announces Antiviral Drug Development Awards, NIH/NIAID News, May 18, 2022

U of M receives NIAID grant to develop drug treatments for future viral pandemics, U of M News, May 23, 2022

Reinforcement Learning with Real-time Docking of 3D Structures to Cover Chemical Space: Mining for Potent SARS-CoV-2 MainProtease Inhibitors, Oct. 5, 2021