According to The Nikkei Asian Review, the Japanese government seeks to have researchers use artificial intelligence to streamline the development of new drugs, aiming to boost the nation's competitive advantage.
The health ministry plans to make such technology available to pharmaceutical companies in a few years. A self-learning AI will be developed starting in fiscal 2017 at the government-affiliated National Institutes of Biomedical Innovation, Health and Nutrition.
The AI will mine Japanese and foreign research papers and databases on new drugs for various conditions. It will then discover candidate compounds, commonly known as seeds, for animal and other testing whose results will add to the AI's knowledge.
The AI will be put to use in a drug discovery support network that includes the Japan Agency for Medical Research and Development, the Riken research institute, and the National Institute of Advanced Industrial Science and Technology. The health ministry will initially pour in 350 million yen (US$3.47 million) during the fiscal year starting April 2017, with budget requests to increase later.
AI is finding use in industries ranging from finance to manufacturing. For medicine, the University of Tokyo has been conducting clinical research with IBM since 2015 where an AI system studies oncology papers and provides assistance in cancer diagnosis. It takes nearly a month for human analysts alone to arrive at results, but it can be done in minutes with AI, professor Arinobu Tojo said.
Scientists look for revolutionary new drug candidates that can act on genes and proteins linked to such conditions as cancer and hepatitis C. But pinpointing promising compounds out of many choices and developing a single new drug is said to require at least a decade and tens of billions of yen or more. Success rates are also low.
Japan is still believed to be lacking in drug discovery capabilities compared with global peers. The health ministry sees a need to support researchers though a stepped-up public-private alliance.
Across the Pacific, AI-assisted pharmaceutical development has already taken off. Massachusetts startup Berg uses the technology to compare and analyze some 14 trillion data points from healthy cells and more than 40 types of cancer tissue, including pancreatic and brain tumors. The results have led to the discovery of a new cancer drug.
California startup Atomwise has found that two existing drugs may also work against the Ebola virus. The company's AI reportedly achieved in only a day a feat that would normally take years. Pharmaceutical spending accounted for a fifth of the 41.5 trillion yen that Japan shelled out for health care last fiscal year. Health insurance programs risk collapsing under the weight of cancer immunotherapy and other pricey new treatments, the medical community warns.
If you want to read this article in Japanese, please see the following link:
Subscribe to our English Newsletter