Google Research and Google DeepMind have developed a new LLM called Tx-LLM for drug discovery and therapeutic development by fine-tuning from PaLM-2. Tx-LLM targets 66 tasks in drug discovery and utilizes PaLM-2 technology. It was trained using 709 datasets and constructs the TxT collection for drug synergy prediction by combining free-text instructions and representations of small molecules. Tx-LLM outperformed state-of-the-art models on 43 tasks and showed positive transfer between datasets. The use of artificial intelligence in drug discovery is a growing trend, with companies like Absci and IBM collaborating with pharmaceutical giants to expedite novel treatments. The HIMSS AI in Healthcare Forum will discuss advancements in AI technology in healthcare.
Source link
Google creates Tx-LLM for drug discovery and therapeutic development
