"Based on the trial results of the 6 patients currently enrolled, the response of virtual patients to the drug is consistent with the results of real clinical trials." At the "Computational Medicine" sub forum of the China (Suzhou) Innovative Pharmaceutical Medicine Conference recently held, the latest research results brought by Jiang Min, Director of the Drug Clinical Trial Office of Peking University Cancer Hospital, sparked the interest of many participants. With the acceleration of generative artificial intelligence applications, the simulation interaction of artificial intelligence provides more new means for scientific discovery. In 2023, in order to promote the deep integration of artificial intelligence and scientific research, the special deployment work of "AI for science" will be launched. Among them, drug development is one of the key areas of deployment work. "Whether drug reactions in the human body can also be accurately presented through digital twins is a scientific question that we need to verify." As a collaborator in the "virtual clinical trial" research with Peking University, Zhao Yu, a researcher at the Western Institute of Computing Technology of China Academy of Sciences and deputy director of Turing Darwin Laboratory, said that using computational medicine technology to establish digital twins and drug models for cancer patients, and then conducting basic theoretical and applied research related to pharmaceutical innovation and development, is one of the important directions for "artificial intelligence driven scientific research". "To make up for the shortcomings of the difficulty in reading massive amounts of data," said Niu Gang, a researcher at the Western Institute of Computing Technology of China and the director of the Turing Darwin Laboratory. "Currently, there is a large amount of scientific literature in the pharmaceutical field, and taking the research on rabies virus as an example, there may be tens of thousands of related literature. It may take several years for a researcher to achieve comprehensive reading." Artificial intelligence can greatly compress this time-consuming accumulation process, not only load all knowledge, but also use technology such as model training and mining to achieve correlation analysis of related knowledge, thereby promoting innovation in the pharmaceutical research and development field. ". Yang Mei, Deputy Chief Physician of the Breast Department at Guangdong Provincial People's Hospital, deeply feels the assistance of artificial intelligence. She started collaborating with the Turing Darwin Laboratory computational medicine team 7 years ago. Based on the new technology system of computational medicine, Myrica rubra led the team to propose an embryonic genome etiology framework, and carried out clinical trials of embryonic genome etiology with female breast cancer as the specific research object; She developed the "Germline Genome Shotgun Damage Assessment" system, which significantly improved the research level of breast cancer etiology in China, and provided a new path to reveal the incidence of breast cancer in China, and to explore breast cancer prevention and screening strategies consistent with Chinese characteristics. It is reported that West China Hospital of Sichuan University has established a knowledge base and artificial intelligence model for epilepsy in women of childbearing age. Chen Lei, Vice Dean of West China Hospital of Sichuan University, introduced that the construction of knowledge bases and artificial intelligence models can be used to guide clinical practice on one hand, and on the other hand, to explore the potential new uses and scope of existing old drugs in epilepsy treatment, explore new medical mechanisms, and provide new methods and ideas for the research team. We are utilizing computer modeling and simulation technology to develop patient specific digital models to form virtual patient groups for testing the safety and effectiveness of new drugs and medical devices, overcoming the long clinical trial cycles and high investment "pain points"