Scientist: Artificial Intelligence+Molecular simulation and drug development prospects are broad

On July 12th, the 02 -issue lecture of "Understanding the Future" in 2022- "AI+molecular simulation and drug research and development" was held on the entire network. Future Forum Director Xie Xiaoliang, a professor at Peking University Li Zhaoji, served as the host, and specially invited Jiang Hualiang, an academician of the Chinese Academy of Sciences, researcher at the Shanghai Institute of Pharmaceuticals of the Chinese Academy of Sciences. The cutting -edge academic achievements, talk about deep integration and innovation in cross -disciplinary cross -disciplinary cross -disciplinary fields. They pointed out that artificial intelligence+molecular simulation and drug research and development potential are huge and have a broad prospect.

In the keynote speech, Professor Jiang Hualiang introduced the progress and trend of international innovative drug research and development based on the title of "AI empowering innovative drug research -status quo and future". application.

He pointed out that the application of AI technology in the research and development of drugs has attracted great attention from the research institute and the pharmaceutical industry, and began to empower drug research and development from target discovery and confirmation, the discovery and optimization of pharmaceuticals, pharmaceuticals and toxic evaluations. It will become one of the key core technologies for future drug development. However, the current application of AI in drug research and development is also in the initial stage. It is necessary to develop new AI technologies for drug development and closely combine with traditional drug molecular design and experimental technology to truly empower drug development. He specifically raised ten challenges used by AI to develop drugs, and encouraged young students to participate in the development of this field.

Professor Gao Yinqin gave a keynote speech based on the "Methods and Applications of Molecular Simulation Combination in Molecular System", which introduced how to use deep learning technology to overcome the bottlenecks of traditional molecular simulation, molecular simulation combination of deep learning methods in protein structure prediction, molecular molecular The application of docking and chemical reaction mechanisms shows the wide application prospects of molecular simulation and deep learning methods, and further look forward to the development direction of future molecular simulation technology.

He pointed out that traditional molecular simulation is severely limited in time and space when applying molecular systems such as complex chemistry and biology. The artificial intelligence technology represented by deep learning can establish an organic connection between theoretical and computing, theory and experiments, computing, and experiments. Therefore, it has become an important tool for breaking the traditional molecular simulation bottleneck, molecular simulation and molecular science.

In the forward -looking dialogue, the guests start a dialogue , Share deep insights and thinking.

Professor Jiang Hualiang pointed out that there are still some key issues in small molecular drugs that need to be resolved in AI. For example, the calculation speed of freedom can be increased by 3-5 times compared with the past. When the speed is increased to tens of times, the combination of free can predict accuracy and the efficiency of small molecular drug design is expected to achieve essential breakthroughs. In his opinion, it is difficult to improve the pain points such as high investment and long cycle in the field of pharmaceuticals, but AI will be promising in predicting the success rate of clinical candidates. "In clinical trials, only one of the 10 candidate drugs has only one kind of test successful test, and we have accumulated clinical data of tens of thousands of drugs, which contains universal data of a large number of previous clinical trials. Forecast and eliminate failure drugs in clinical candidates, and better lock in drugs that may be successful in clinical trials. "

Professor Gao Yiqin said that data is the biggest bottleneck that restricts the design of small molecular drug design. "At present, there are very few reliable data that can be truly obtained. There are still problems such as inconsistent indicators and difficulty obtaining indicators in data. Increasing high -precision calculations are also enhanced by data. The ground is coupled to form a seamless connection closed loop, and the possibility of obtaining high -quality data will be greatly improved. "In addition, Professor Gao Yiqin also pointed out that by integrating single -cell group information and establishing a reliable cell response model, AI can make AI pairs in pairs. Drug research and development of downstream to make some pre -judgment work. "If the flux is high enough, the cell model can be used to predict the small molecular access, protein signal transmission, and protein entering of small molecular drug design, as well as the large molecular drug design. Self -learning and optimization, the accuracy of prediction will gradually improve. If it is made into a open platform for publicity, it will benefit the entire pharmaceutical research and development. "

Professor Xie Xiaoliang also shared the thinking of AI in the field of drug research and development in the forward -looking dialogue. He pointed out that small molecular drugs are combined with macromolecules. At present, the sequences and structural data of macromolecular are large. The existing technology has entered the name of protein on the computer terminal. About %, but this is the ideal situation at present. Small molecules are combined on the protein macromolecules, and the AI ​​and MD models are combined with algorithms. At present cost. However, due to the insufficient amount of small molecules, the database is not large enough, and the machine learning prediction of small molecular drugs cannot be achieved. This is a huge challenge faced by small molecular drug design. (Guangming Daily All -Media Reporter Yuan Yufei)

- END -