Challenge
The Nation possesses the biomedical knowledge and clinical infrastructure to identify therapeutic relevance faster than it can turn these discoveries into treatments. Bringing new drug therapies to patients can still take over a decade and cost billions of dollars as many candidates fail late. This challenge intends to reduce the timeline by half in ten years, doubling the number of drug-based treatments we bring to patients as a nation. For example, in rare diseases, which impact 30 million Americans, 95% of 10,000 have no viable treatment options—small market size and the number of disparate conditions create economic barriers that standard drug development models cannot overcome. Thousands of approved and investigational drugs may hold untapped therapeutic value, yet the genomic, chemical, clinical, structural, and real-world evidence needed to find those uses sits in disconnected systems that cannot be queried together.
AI Solution
Researchers will leverage AI with integrated molecular, genomic, proteomic, phenotypic, clinical, and real-world evidence data sources from government and private partners to identify previously unrecognized relationships between drugs, targets, diseases, patient populations, and new surrogate endpoints. By aligning public-private datasets with predictive design models, the Nation can shift towards broad-spectrum, platform-based therapeutic discovery encompassing both repurposing of existing drugs and generation of new ones.
Improved ability to evaluate drug and therapeutic candidates across new data sources will accelerate discovery by prioritizing safety profiles, biological plausibility, or potential to treat multiple indications simultaneously. Advanced capabilities running on DOE high-performance computing infrastructure and the DOW’s high throughput screening and specialized validation infrastructure will enable rapid evaluation of therapeutic mechanisms, predict the effect of target engagement, and treatment-response predictions across diverse biological contexts.
Autonomous laboratory platforms will be used to experimentally validate high-priority hypotheses while longitudinal cohorts, and clinical trial networks will provide downstream validation of repurposing candidates most likely to generate meaningful clinical and national security impact.
This will substantially compress the time required to identify, evaluate, and validate new uses for existing therapies while improving the efficiency of clinical development.
Justification
AI will enable the Nation to rapidly identify high-confidence drug repurposing opportunities. Through leveraging the Department of Health and Human Services’ biomedical knowledge, spanning decades of therapeutic development, translational pathways, and clinical trial networks, the Department of Energy's exascale computing resources, and the Department of War’s end-use driven research, drug researchers will optimize development pipelines, thereby shortening the pathway from scientific discovery to patient benefit.
National Impact
Creating a scalable national infrastructure that continuously converts biomedical knowledge into actionable therapeutic opportunities can establish the United States as the global leader in AI-enabled translational science. Success would benefit patients across a wide range of diseases, including rare diseases, cancer, neurodegenerative disorders, infectious diseases, and chronic conditions, while providing a reusable framework for future biomedical discovery initiatives. The resulting capabilities would also strengthen national competitiveness in biotechnology, AI-enabled medicine, and precision health, potentially doubling the productivity and impact of American science and innovation.
Challenge
This was last updated .