AI Model TxGNN Repurposes Drugs for 17,080 Rare Diseases

Sep 25, 2024, 03:30 PM

Researchers have developed TxGNN, an AI-driven model designed to repurpose existing drugs for the treatment of rare diseases. This model, recently published in Nature Medicine, identifies potential therapeutic candidates for over 17,080 diseases, including those with no current treatment options. TxGNN leverages a medical knowledge graph to rank drugs as potential indications and contraindications, offering new hope for approximately 300 million people worldwide affected by rare diseases. The model's zero-shot framework allows it to predict drug candidates without prior examples, significantly accelerating the drug discovery process. TxGNN was trained on data from 7,957 drugs.