Accelerating Drug Discovery: AI-Powered Screening of 100 Million Compounds Daily with ChatGPT

Researchers from MIT and Tufts University have developed a computational approach to accelerate drug discovery by training a large language model to analyze protein-drug interactions. The model can match target proteins with potential drug molecules without the need to calculate the molecules’ structures first, reducing the time required to screen candidate drugs. The researchers claim that they can screen more than 100 million compounds in a single day using their method. Results from experiments showed that 12 out of the 19 drug-protein pairs chosen from the top hits had a strong binding affinity, while nearly all of the possible drug-protein pairs would have no affinity. The researchers plan to apply this approach to other types of drugs, such as therapeutic antibodies.