Google DeepMind accelerates discovery of liver disease mechanisms with AI Co-Scientist

Bioengineer Filippo Menolascina at the University of Edinburgh used Google DeepMind's Co-Scientist to comb scientific literature for overlooked links in metabolic dysfunction-associated steatohepatitis (MASH). MASH involves intertwined processes like liver inflammation and metabolism, making single-target drugs insufficient. Co-Scientist narrowed the search for combination therapies by synthesizing evidence across liver biology and pharmacology. In a key case, it tackled why the drug resmetirom benefits only a narrow patient group, hypothesizing that the NLRP3 inflammasome bridges inflammation and metabolism in the disease—a connection not previously assembled. The hypothesis was later experimentally verified, potentially guiding targeted dual-therapies.