Stanford Medicine and Calico Life Sciences Deploy DeepMind Co-Scientist AI to Target Liver Fibrosis and Aging

Biomedical investigators across the San Francisco Peninsula and East Bay life sciences corridor are accelerating therapeutic discovery using Google DeepMind's Co-Scientist artificial intelligence system. At Stanford University School of Medicine along Campus Drive, geneticist Gary Peltz deployed the tool to identify candidate drugs capable of halting liver fibrosis, which causes more than 1.4 million deaths each year. Published in Advanced Science on Saturday, October 17, the research program shows how computational hypothesis generation can repurpose existing medicines for patients in San Francisco, Walnut Creek, and Contra Costa County.

To test the system, Peltz asked Co-Scientist to propose three candidate drugs from existing medicines while also identifying two candidates himself based on their presence in liver fibrosis literature. In lab tests using live human liver cells, two of the three drugs selected by Co-Scientist blocked fibrosis and promoted regeneration of liver cells. Meanwhile, Peltz's two manual drug picks showed no benefit against fibrosis during the controlled laboratory tests.

The standout artificial intelligence pick, the cancer drug vorinostat, blocked 91 percent of a damage response that can drive liver scarring. Regional biotechnology board members and hospital department officials noted that repurposing approved compounds can save over $50 million in early screening costs. Hepatology specialists serving medical centers in Oakland, San Ramon, and Palo Alto emphasized that blocking 91 percent of fibrotic signaling marks a major milestone for regenerative medicine.

Further north along Oyster Point Boulevard in South San Francisco, Calico Life Sciences researchers Matt Onsum and Katherine Labbé are using Co-Scientist to connect scattered findings across the biology of aging and turn them into hypotheses worth testing. Onsum said Co-Scientist impressed Calico's experts and scientific committee with its scientific discernment, cutting through noise to generate a novel yet plausible hypothesis about how the integrated stress response (ISR) is regulated by metabolism. Researchers interacted with Co-Scientist to refine the experimental design and feed in new information as results emerged, leading to important new findings on ISR in health and disease that the team plans to publish.

Academic review boards, county health district officials, and biotechnology council leaders across the Bay Area expect AI drug discovery partnerships to expand this autumn. Investigators can review the findings online at med.stanford.edu or contact the Stanford Medicine research office at 279 Campus Drive by calling (650) 723-4000 between 8:00 AM and 5:00 PM Monday through Friday. Additional aging research updates are available through calicolabs.com.