50 companies in Japan want to use artificial intelligence to develop new drugs

A coalition of companies and research institutions will be launched soon. Not only in Japan, but also overseas IT companies and pharmaceutical companies are expected to participate. Strive to develop new drug development based on artificial intelligence with the goal of three years later. The Ministry of Education, Culture, Sports, Science and Technology will add 2.5 billion yen to the 2017 budget estimate request to support the project. The final total is expected to reach 10 billion yen.

New drug development must first discover new drug candidate substances. These substances act on proteins and the like which are involved in the onset of the disease. It is then necessary to verify the effectiveness and safety of the animal and the human body. The realization of productization is considered to take more than 10 years and more than 100 billion yen. However, there are many situations in which problems are found in the final stage and have to be suspended. This will result in the pharmaceutical companies being unable to recover their investment, bearing a huge burden and causing performance deterioration.

The companies and research institutions involved in the project will send researchers to form a team of 100 people to promote development. Artificial intelligence will read the data and learn, then give the answer. In addition to the academic papers related to the target new drug, the patient will also input artificial clinical data, disease-related proteins, and new drugs locked by the world's few supercomputers "Kyo", which are held by the Institute of Physical Chemistry and Kyoto University Hospital. Information such as candidate substances, thereby discovering new drug candidate substances. The artificial intelligence developed will be brought back by each pharmaceutical company to find new drug candidate substances.

By using artificial intelligence, it is possible to significantly shorten the search time for new drug candidates that would otherwise take 2 to 3 years. If artificial intelligence can eliminate new drug candidates that may have side effects, it is expected to increase the success rate of new drug development. In addition, it is also possible to find cheap substances with the same effect as high-priced drugs, which will help to significantly reduce the price of drugs.

The trend of trying to develop new drugs through artificial intelligence is expanding. New start-ups in the United States are under development. In addition, research is under way to promote the use of computerized "Watson" from IBM in the United States for the diagnosis of diseases. In Japan, the National Research and Development Corporation's Institute of Basic Health and Nutrition is planning to use the existing artificial intelligence to find new drug candidates from 2017. It is reported that the newly developed artificial intelligence for the development of new drugs can improve the accuracy by learning various data, and can also track the causes of the effects of new drug candidate substances, which is helpful for promoting clinical application.

Takeda, Japan's largest pharmaceutical company, ranks 17th in the global pharmaceutical industry, and is inferior in size to pharmaceutical giants such as Pfizer and Swiss Novartis. Development costs are only less than half of the major pharmaceutical companies such as Pfizer. Without artificial intelligence to improve R&D efficiency, Japanese pharmaceutical companies will not be able to win in global competition.