Fujitsu Collaborates with Riken to Harness Generative AI for Accelerating Drug Development
Fujitsu partners with Riken to develop generative AI technology for drug development. Advanced cryo-electron microscopy used to capture protein images. Generative AI technology reproduces proteins as three-dimensional structures in motion.
In a groundbreaking partnership, Fujitsu has joined forces with the government-backed Riken research institute to develop generative AI technology that has the potential to revolutionize the field of drug development. This cutting-edge technology is capable of predicting the state of a drug's target protein in the body at a speed more than 10 times faster than existing methods.
The collaboration between Fujitsu and Riken involves utilising advanced cryo-electron microscopy to capture images of proteins. These images are then processed using generative artificial intelligence, which can accurately reproduce the proteins as three-dimensional structures in motion.
Trials for this innovative technology are set to commence in fiscal year 2024. Fujitsu has already reached out to approximately 90 members of the Life Intelligence Consortium, which includes pharmaceutical companies and universities, to participate in these trials.
The administration of drugs often relies on substances that bind to specific proteins responsible for causing diseases or other medical conditions. By accurately estimating the shape and movements of the targeted protein, this new technology has the potential to significantly speed up the drug development process. In fact, in an experiment using ribosomal protein, a task that previously took specialists an entire day was reduced to just two hours.
Traditionally, researchers have relied on analysing a large number of electron microscope images to predict the three-dimensional structure of a protein. However, accurately predicting its movements has proven to be a challenging task. Due to the limitations of what can be examined at once, this process can take months to complete.
The ultimate goal is to develop AI technology that can predict the movements of proteins at the atomic level, based on genetic material. Takashi Kato, from Fujitsu's Artificial Intelligence Laboratory, believes that this breakthrough could be applied in the drug discovery field within the next five years.
The use of generative AI for detailed analysis in the early stages of drug development would allow resources to be focused on the most promising candidates. This targeted approach increases the probability of successful drug development and could potentially lead to faster treatments for infectious diseases.
According to the McKinsey Global Institute, the economic impact of generative AI in the pharmaceutical and medical products industry, including reduced development costs and increased sales of new drugs, is estimated to be between $60 billion and $110 billion annually.
Fujitsu partners with Riken to develop generative AI technology for drug development
Advanced cryo-electron microscopy used to capture protein images
Generative AI technology reproduces proteins as three-dimensional structures in motion
Source: NIKKEI ASIA