AI Agents Discover Anti-Aging Drugs by Mining 2 Million Molecular Profiles
Autonomous AI system reanalyzed millions of studies to find 500+ interventions that reduce biological age, validating ouabain in mice.
Summary
Researchers developed ClockBase Agent, an AI system that autonomously reanalyzed 2 million human and mouse molecular profiles from decades of research using 40+ aging clocks. The AI discovered over 500 interventions that significantly reduce biological age—effects missed by original investigators who weren't studying aging. The system identified ouabain, a cardiac drug, as a top anti-aging candidate and validated it in aged mice, showing reduced frailty, improved heart function, and decreased brain inflammation. This represents a new paradigm where AI systematically mines all past research to discover longevity interventions.
Detailed Summary
Scientists have created the first AI system capable of autonomously discovering anti-aging interventions by reanalyzing decades of molecular research. ClockBase Agent processed 2 million human and mouse samples from public databases, applying over 40 aging clocks to identify biological age effects that original researchers never looked for.
The AI analyzed 43,602 intervention-control comparisons across genetic modifications, drugs, environmental exposures, and disease models. It discovered 5,756 significant age-modifying effects, including over 500 interventions that reduce biological age. Top candidates included ouabain (a cardiac glycoside), KMO inhibitors, fenofibrate, and various genetic knockouts.
Key patterns emerged: significantly more interventions accelerate aging than slow it, disease states predominantly increase biological age, and genetic loss-of-function approaches outperform gain-of-function strategies for anti-aging effects. The identified interventions converged on known longevity pathways like mTOR, autophagy, and cellular senescence.
To validate their approach, researchers tested ouabain—a cardiac drug identified by the AI but never studied for anti-aging effects. In aged mice, ouabain treatment reduced frailty progression, improved cardiac function, and decreased neuroinflammation, confirming the AI's prediction.
This work establishes a new paradigm where specialized AI agents systematically reanalyze all historical research to extract aging-relevant insights. By applying standardized aging biomarkers to experiments never designed to test longevity, the system transforms the entire archive of molecular research into an aging intervention discovery engine, potentially accelerating the identification of compounds that could extend human healthspan.
Key Findings
- AI system identified 500+ interventions that significantly reduce biological age from existing data
- Ouabain treatment reduced frailty and improved cardiac function in aged mice
- Loss-of-function genetic approaches consistently outperform gain-of-function for anti-aging
- Disease states predominantly accelerate biological aging across all datasets
- More interventions accelerate aging than slow it, revealing fundamental biological constraints
Methodology
Researchers processed 2 million samples using 40+ aging clocks, then deployed AI agents to autonomously analyze 43,602 intervention-control comparisons. The system generated hypotheses, performed statistical analyses, and conducted literature reviews to identify age-modifying effects with composite scoring for prioritization.
Study Limitations
This is a preprint study requiring peer review. The AI analysis relies on transcriptomic aging clocks which may not capture all aspects of biological aging. Experimental validation was limited to one compound (ouabain) in mice, and human translation remains to be demonstrated.
Enjoyed this summary?
Get the latest longevity research delivered to your inbox every week.