A Survey of Large Language Model for Drug Research and Development

作者：Huijie Guo, Xudong Xing, Yongjie Zhou, Wenjiao Jiang, Xiaoyi Chen, Ting Wang, Zixuan Jiang, Yibing Wang, Junyan Hou, Yukun Jiang, Jianzhen Xu · 发表于：IEEE Access · 年份：2025 · DOI：10.1109/access.2025.3552256 · 被引用次数：12 · 研究领域：Topic Modeling

Drug research and development (drug R&D) is a sophisticated, cost-intensive, and time-consuming procedure with historically low success rates. The advent of Artificial Intelligence (AI) technologies has introduced innovative methods into drug R&D, particularly by leveraging AI capabilities. Large language models (LLMs), a breakthrough in generative AI, have revolutionized drug discovery. With their extensive datasets, numerous parameters, and strong multitasking abilities, LLMs have significantly improved efficiency across various related domains, providing unparalleled support to drug R&D. These models have facilitated a deeper understanding of intricate disease mechanisms and the identification of novel therapeutic strategies, ushering in a new era in drug development and clinical applications. As a result, the advancement of LLMs is poised to drive significant transformations in drug R&D, emphasizing the importance of effectively leveraging this technology. This review provides insights into the architecture and characteristics of LLMs, explores their applications in drug R&D, and highlights their research implications in bioinformatics data, including proteins, genes, and chemical compounds. Furthermore, it investigates the practical strategies of LLMs in drug discovery, drug repositioning, and clinical inquiries, presenting an innovative approach to research and future advancements in this field.