Computing medicine for the pharmaceutical industry "Diaoyun See the Day"

The dividend of traditional biochemical technology in the field of medicine seems to have come to an end.

At present, the cost of research and development of a new drug is about $ 2.6 billion, which takes about 10 years, but the success rate of the approval of drugs (LOA) during the clinical trial stage is less than 10%, especially the tumor drugs are only 5.1%. This means that many new tumor new drug research and development projects cannot be separated from the fate of being eliminated.

In view of this, whether it is the government's planning documents, or the frequent actions of investment and financing institutions and pharmaceutical companies, the innovation motivation of the pharmaceutical industry hopes to empower the information technology.

In January 2022, the "Fourteenth Five -Year Plan" Digital Economy Development Plan issued by the State Council proposed that "the strategic choice of developing a new round of scientific and technological revolution and new opportunities for industrial changes", "Promoting digital technology to economic, social and industrial development Wide and in -depth penetration in various fields. "

Subsequently, the "Fourteenth Five -Year Pharmaceutical Industry Development Plan" jointly released by the Ministry of Industry and Information Technology, which is jointly released by the Ministry of Industry and Information Technology, clearly stated that "the new generation of information technology empower pharmaceutical research and development. Explore artificial intelligence, cloud computing, big data, etc. The application of technology in the field of R & D, which can improve the efficiency of new targets and new drugs through analysis and simulation computing of biological data. "

So, what are the aspects of information technology empowering the biomedical industry? To this end, the reporter interviewed experts in related fields.

The status quo of innovation

At present, the biomedical industry has become an important force leading China's economic development. But unfortunately, there is no "Chinese -made" medicine in the global TOP100 sales.

Zhang Chunming, an associate researcher at the Institute of Calculation of the Chinese Academy of Sciences and executive deputy dean of the Chinese Academy of Science and Technology, pointed out that domestic innovative drugs are mostly concentrated abroad. Homogeneous competition.

For example, in the target of type 1 small molecular anti -tumor drugs, the first three target EGFR, CDK4/6, and ALK account for 34%.

In this regard, Zhang Chunming bluntly stated that if according to such a "volume method", each pharmaceutical company cannot make money, and the industry will fall into a vicious circle in the end.

In addition, tumors, cardiovascular diseases, and Alzheimer's diseases that seriously threaten human health are caused by most of them, and it is difficult to be effective for all patients by only one drug or standardized treatment plan.

"The human disease model is still based on model animals. But there is a large difference in the process, pathological mechanism, status, and recovery after healing of animal disease models. Cognition is still in the stage of local or fragmented, and some new discoveries are still in the "blind people touching '." Cong Bin, an academician of the Chinese Academy of Engineering, believes that understanding the law of people's life activities based on cells and animal models has caused the research and development of medicine and medicine. An important reason for the predicament.

New ideas brought about by calculation

Indeed, the biological system itself is very complicated, coupled with the multi -dimensional nature of the tumor pathogenesis, it directly exacerbates the difficulty of the development of new drugs.

Fortunately, with the in -depth advancement of technologies such as big data, Internet, biological information, and artificial intelligence (AI) in the field of biomedicine, the model of new drug development has gradually changed.

Cong Bin pointed out that in the era of digital economy, if you can build the digitalization of the biological world and realize the digital twins of life functions, to maximize the law of human life and disease model, it will inevitably provide new ideas for the development of new drugs.

This is also recognized by experts at home and abroad.

At the end of 2020, after investigating more than 90 top cancer experts after investigating more than 90 topic cancer experts, it was published in the CA CANCER J Clin journal that the calculation method and mathematical model were detected to detect the evolution of tumors and predicting the treatment of drug resistance. Powerful means of clinical impact.

Liu Changxiao, an academician of the Chinese Academy of Engineering, also publicly wrote that the combination of AI and drug research and development is applied to discovering drug targets, excavating candidate drugs, high -throughput screening, drug design, drug synthesis, predicting drug absorption, allocation, metabolism, excretion and toxicity The prospects of the development of new indications are bright. Among them, target screening is the hottest area of ​​recent AI+drug development, and drug synthesis may move towards automation.

The advantage of calculating medicine

However, with big data, it does not mean that it is "omnipotent" in the field of pharmaceutical research and development, but also requires the support of algorithms and computing power.

To this end, the high -performance computing team of the Institute of Computing Technology of the Chinese Academy of Sciences proposed the concept of computing medicine. The so -called computing medicine is based on systematic theory, adopts dense data drive as a scientific research paradigm, with artificial intelligence as the method, high -performance calculation as support, through the double -wheel drive of knowledge model+data model, as a full chain in the field of biomedicine New insights and solutions for industrial contributions.

Specifically, there are four characteristics of computing medicine: one uses the overall theory of complex system science as a way of thinking, and understands that biomolecules, cells, tissue organs, population and other biological level structures New attributes to capture the mechanism of the disease from the perspective of the system; the second uses a scientific paradigm driven by dense data to tap the new hole in biomedical big data hidden in high -dimensional, high -throughput, and multi -dimensional fusion; Convert the knowledge model in the field of biomedicine to a mathematical model, use the big data of biomedical as the input parameter, iterate and train the model with artificial intelligence algorithm, and the output is close to the real life system structure and functional characteristics, so as to understand the nature of the disease occurred ; Finally, use high -performance calculation as a new generation of computing infrastructure, and provide data storage, calculation accuracy and computing speed support for new scientific discovery. "Computing the mechanism results provided by the medical platform can directly simulate the actual efficacy of the drug under the conditions of clinical trials, which can seamlessly continue biological experiments and clinical trials. The success rate of phase II/III clinical trials. "Said Niu Gang, a guest professor professor of the Western Research Institute and the director of the Turing Darwin Laboratory of the Western Research Institute of China Science and Technology.

Policy frequently releases signal

The knowledge system for calculating medicine also meets the essentials of the current policy documents.

Globally, the United States has launched the application of AI technology in the field of pharmaceutical health earlier in the United States. AI R & D investment and management.

In contrast, in my country, in November 2021, the "Principles of Clinical R & D Guidance of Anti -tumor Drugs, guided by clinical value" released by the China National Drug Administration (CDE), specially emphasized the "strengthening mechanism research" and started at the beginning of the drug research and development. Basic research on tumor occurrence and development mechanism should be strengthened, and the research of drug action mechanism should be strengthened. Through breakthroughs and innovation, optimize drug design, develop new treatment methods, and continuously meet the treatment needs of tumor patients. In addition, the above -mentioned "Guidance Principles" also emphasizes "improving precision treatment", which should pay attention to and continuously improve the degree of matching patients and therapeutic drugs, and continuously improve the precision of anti -tumor treatment, so that tumor patients can receive more suitable treatment for themselves drug.

The "Fourteenth Five -Year Plan" Biological Economic Development Plan also emphasizes relying on artificial intelligence technology, biomedical and healthy big data resources, developing intelligent auxiliary decision -making knowledge models and algorithms, auxiliary personalized new drug research and development, and providing decision -making support for disease diagnosis and treatment.

Industrial layout is full of drums

The introduction of policies is not only a signal of industrial development, but also outlines the blueprint for the industrial layout.

In 2019, Lilly cooperated with Atomwise to use the latter's AI drug exploration technology to accelerate drug screening. Sanofi cooperated with Novadiscove in 2021 to use the latter's advantage to laid out the disease model in virtual patients to support decision -making and reduce clinical development risks. Subsequently The ability of artificial intelligence and personalized pharmaceutical platforms, screening small molecular candidate drugs in the field of tumors and immune. In addition, Pfizer also restarted Cytoreason cooperation to try to apply the latter's artificial intelligence technology to Pfizer's drug development.

In the field of investment and financing, in March 2021, AI Pharmaceutical Company Insitro announced the completion of $ 400 million in Series C financing, becoming a company with the highest single financing in the world's AI pharmaceutical field. Recently, a domestic Institute of Calculation Technology Institute of the Chinese Academy of Sciences has been incubated by Zheyuan Technology, which is committed to using AI, big data, and biomedicine to empower innovative drugs.

It is reported that Zheyuan Technology has set up two AI tool platforms -the new target and new mechanism delivery platform, virtual clinical trial platform, pointing to six application scenarios: new targets, new action mechanisms, new biomarkers, design clinical clinical clinical clinical clinicals Test schemes, predict clinical trial results, and expand new indications.

The new base technology finally "see the clouds"

"Machine learning is a brand new cognitive methodology that cannot really understand but can be tested by practice." Li Guojie, an academician of the Chinese Academy of Engineering, pointed out that the cross -border challenges of artificial intelligence and new drugs are not only technical, but also technically. More importantly, whether the mode of thinking and understanding of knowledge can be successfully connected.

At present, my country's big data and artificial intelligence technology are in the global "running" position. However, the practice of data technology and biomedical industry is backward.

Zhao Yu, deputy director of Turing -Darwin Laboratory and Zheyuan Technology COO, believes that from a global perspective, although the application of artificial intelligence and big data in biomedicine is still in the early stage, based on the biomedical industry in my country The critical climbing period of improving innovation, calculating medicine as a new technical system, a drug digital test field, is bound to "take a different approach" for new drug research, and it can also "set the clouds" for the current biomedical industry.

The understanding of life and disease is the source of biomedical innovation.Zhang Chunming believes that in the era of digitalization of life sciences, Zheyuan Technology is committed to establishing a new base technology that cognates life, becoming an innovative factory for drug IP, bringing more innovative results to the industry, improving drug research and development efficiency, and for drug research and developmentIntroduce a new paradigm to provide patients with more effective treatment plans.(Guangming Daily All -Media Reporter Cui Xingyi)

- END -