Large Language Models in Drug Discovery: A Survey

Drug Discovery is a very lengthy and resource-consuming process. However, a variety of advanced Artificial Intelligence (AI) and Deep Learning (DL) techniques are being utilized to accelerate and advance DD, such as Large Language Models (LLMs). This survey is in aim of discovering and comparing the currently available LLMs, their methodologies, used datasets, and the different tasks they are aiding in in the DD process, in particular; de novo drug design, drugtarget interaction prediction, masked language models, variational auto encoders, binding affinity prediction, drug repurposing, molecular optimization, activity prediction, contrastive learning for drug-target interaction prediction, and other miscellaneous models. This survey gives insights into future directions and potential in this area.