[MedTech]+[Product]: AI Drug Repurposing Platform DrugSwap Shortens New Indication Discovery from Years to Weeks
The DrugSwap platform, jointly developed by Harvard University and Insilico Medicine, uses graph neural networks to analyze molecular structures and disease target relationships of existing drugs, enabling rapid discovery of new uses for old drugs.
Article Body
New drug development averages 12 years and $2.6 billion in costs, while drug repurposing can reduce this timeline to 2-3 years. The DrugSwap platform, jointly developed by Harvard University's Department of Computational Biology and Insilico Medicine, is further accelerating this process to just weeks.
DrugSwap's core is a molecular interaction map containing over 12,000 marketed drugs and 8,000 disease targets. The platform uses graph neural networks to perform link prediction on this map, identifying potential therapeutic relationships between existing drugs and uncovered disease targets.
In internal validation, DrugSwap successfully predicted known drug repurposing cases with an accuracy rate of 78%. The platform has identified over 200 high-confidence new drug-target pairs, with 12 already entering preclinical validation. DrugSwap expects to release a public academic version by the end of 2030, with enterprise pricing to be determined.
Disclaimer
Content is AI-generated. Do not use it as a basis for real decisions. Do not cite it as factual reporting.