AITHYRA-EPFL wins Google.org funding for virus-guided drug research

An AITHYRA-EPFL consortium has secured $2.4 million over three years through the Google.org Impact Challenge: AI for Science. The project combines generative machine learning with laboratory experiments to identify cellular vulnerabilities that could guide drug design for diseases that are difficult to treat.

Current therapeutic approaches can modulate only 5-10% of the human proteome, leaving most disease-relevant proteins beyond the reach of conventional medicines. The researchers plan to use viral proteins as guides because viruses have evolved to target critical regulatory sites and hijack cellular machinery. Generative models will map virus-host interaction sites, while high-throughput functional assays will test their cellular relevance.

The team brings together AITHYRA researchers Michael Bronstein, Jason Nomburg and Georg Winter with EPFL scientists Bruno Correia and Nicolas Thomä. Its goal is to turn insights from viral evolution into predictive drug design rather than relying on chance discoveries.

The consortium is among 17 organizations selected globally from more than 3,000 applications across 30 countries to share an updated funding total of $36 million. Recipients may also participate in a Google.org Accelerator, with six months of dedicated pro bono technical support from Google experts and access to Google Cloud credits.