OpenAI: Using AI agents to uncover new leads in biological data

OpenAI has developed a method using AI agents to analyze complex biological data, such as protein sequences, and identify new leads for research. Their approach uses reinforcement learning from human feedback to develop a model that can predict protein-ligand binding affinity, which is crucial for designing new medicines. The model was trained on a dataset of 1.7 million protein sequences and 2.5 million ligands. The AI agents were able to identify new targets for existing drugs, as well as new potential drugs for certain diseases. This approach has the potential to accelerate the discovery of new medicines and treatments.