Artificial intelligence may promote the development of innovative drugs in China to achieve leapfrog progress

[pharmaceutical network market analysis] some insiders said that the past five years have witnessed the vigorous development of innovative drugs in China It is understood that under the promotion of a series of policies, such as the acceleration of the review and approval of new drugs, "4 + 7" volume procurement, the adjustment of medical insurance catalog, and other policies, pharmaceutical enterprises have flocked to innovative drugs and transformed to innovative R & D According to the data that artificial intelligence may promote the leapfrog progress of innovative drug research and development in China (photo source: pharmaceutical network), in 2018, the drug audit center approved 9 domestic class 1 innovative drugs, a new record In recent years, there are many representative products in the approved domestic innovative drugs, such as three domestic PD-1 monoclonal antibodies, the new drug losartan, the first in class drug benvimoder independently developed by China, etc For the rapid development of innovative drugs in China, some analysts pointed out that, on the one hand, the frequent emergence of policies laid a policy foundation for the development of innovative drug industry; on the other hand, the continuous opening of Hong Kong stocks and science and technology innovation board has brought more capital sources for domestic innovative drug enterprises Nowadays, innovative drugs will become a new development point in the field of medicine, and the field of innovative drugs in China has changed a lot However, it is undeniable that innovative drugs have high R & D cost, long R & D cycle and low R & D success rate, which has always been the "three big mountains" of pharmaceutical enterprises, and the vast majority of pharmaceutical enterprises can not achieve profits However, it is pointed out that with the development of artificial intelligence and other technologies, China's innovative drugs may be able to achieve leapfrog development It is understood that artificial intelligence has been involved in drug development For example, the use of artificial intelligence can reduce the overall time-consuming of drug development as much as possible, including automatic film reading, accelerated drug design, virtual high-throughput screening, chip modeling and even later medical simulation and toxicity analysis From the perspective of screening, there has been virtual high-throughput screening At present, artificial intelligence is mainly used in target discovery, compound synthesis, compound screening, crystal prediction, patient recruitment, optimization of clinical trial design and drug indication development and other scenarios in the field of new drug research and development A report from tech energy looked at AI applications in all industries, and the results showed that AI could increase the success rate of new drug research and development from 12% to 14%, saving billions of dollars for the biopharmaceutical industry Drug research and development is one of the important scenarios in the application of artificial intelligence technology Traditional drug research and development is time-consuming and labor-consuming, and the success rate is low AI can greatly shorten the time of drug research and development, improve the efficiency of research and development and control the cost of research and development For drug research and development workers, how to identify the massive scientific research information generated every day is a headache Artificial intelligence technology can extract the knowledge that can promote drug development from these scattered and disordered mass information, and put forward new hypothesis that can be verified, so as to accelerate the process of drug development However, some analysts also pointed out that although artificial intelligence has made great progress in the field of drug development, it can not replace the role of human beings The performance of AI is limited by the quality of input data At this year's Wuxi apptec Global Forum, industry experts also said that many times, high-quality data used to train machine learning algorithms does not exist Low quality data may lead to the algorithm getting false signals from the data, leading to the expansion of "garbage in, garbage out" Today's public data is often unable to become high-quality data due to format, representativeness and other reasons For some AI drug R & D companies, the solution to this problem is to generate high-quality data by themselves For example, insitro and recursion are both generating their own data according to the needs of solving specific problems.