AI-Based Drug Discovery Model Hunts for Key Molecule to Block COVID-19

(San Francisco, CA, USA).

By using predictive models and AI, Atomwise which uses AI for small molecule drug discovery, and its collaborators seek to raise the probability of success for future therapies. Each collaborative project will develop drug candidates with demonstrable broad-spectrum capability, providing potential long-term benefit for future coronavirus outbreaks. Overall, the 15 global research efforts span a wide variety of approaches, including different mechanisms of action, a mixture of viral and human/host target proteins, and targeting conserved regions of proteins that may be recognized even in mutated strains. In addition, many proteins targeted by the academic collaborators have previously been deemed “undruggable” due to their unknown structure or involvement in complex protein-protein interactions. Taken together, the combination of novel approaches could expand the repertoire of therapeutic approaches available for a future outbreak.

Several projects are part of Atomwise’s Artificial Intelligence Molecular Screen (AIMS) program, which enables researchers to accelerate the translation of their research into treatments. In support of each collaboration, scientists at Atomwise will use AtomNet, the company’s patented AI screening technology, to predict the binding of millions or billions of small molecules to a protein of interest identified by the academic researcher as a potential target for COVID-19, narrowing down to a few hundred predicted hit molecules. Atomwise then sources and ships a subset of these predicted compounds to partnering laboratories for testing biochemical potency and selectivity, advancing the most promising compounds for further development as drug candidates.

“Atomwise’s AI screening technology is used to predict the binding of more than 10 million small molecules to a protein of interest, and far exceeds what could be accomplished through traditional laboratory screening methods,” said Dr. Stacie Calad-Thomson, vice president and head of Artificial Intelligence Molecular Screen (AIMS) Partnerships at Atomwise. “With Atomwise’s AIMS Awards program, our hope is to democratize access to AI during the early stages of preclinical drug development and enable academics to contribute to the pandemic response who might not have the opportunity otherwise.”

Related Links:
Atomwise Inc.