AI Co-Scientist Uncovers Liver Disease Mechanism
Researchers at the University of Edinburgh, led by bioengineer Filippo Menolascina, used Google DeepMind's Co-Scientist AI to accelerate discovery of mechanisms behind metabolic dysfunction-associated steatohepatitis (MASH). The AI synthesized evidence across liver biology and pharmacology, narrowing down potential combination therapies from an overwhelming number of possibilities. In a key case, Co-Scientist generated a hypothesis linking the NLRP3 inflammasome to the coupling of inflammation and metabolism in MASH—a connection not previously identified. This hypothesis was later experimentally verified, potentially paving the way for targeted dual-therapies. Menolascina described Co-Scientist as "a jetpack for scientists," predicting a scientific revolution that will shorten iteration cycles for breakthroughs.
Key facts
- Filippo Menolascina is a bioengineer at the University of Edinburgh.
- Co-Scientist is an AI tool developed by Google DeepMind.
- MASH stands for metabolic dysfunction-associated steatohepatitis.
- Co-Scientist identified the NLRP3 inflammasome as a molecular bridge in MASH.
- The hypothesis was experimentally verified.
- Resmetirom is a drug approved for a specific stage of MASH.
- Co-Scientist was used to narrow down potential combination therapies.
- Menolascina believes Co-Scientist will shorten breakthrough iteration cycles.
Entities
Institutions
- University of Edinburgh
- Google DeepMind
Locations
- Edinburgh
- United Kingdom