Fujitsu, in partnership with the esteemed Riken research institute, has embarked on an innovative project to develop generative artificial intelligence technology. This cutting-edge AI is designed to predict the state of a drug’s target protein in the body, achieving speeds over ten times faster than current methodologies. Utilizing high-resolution images obtained via advanced cryo-electron microscopy, the collaboration aims to replicate proteins as dynamic three-dimensional structures.
Starting fiscal 2024, trials are set to commence, with Fujitsu proactively engaging the Life Intelligence Consortium’s diverse members, including key players in the pharmaceutical industry and academia. This consortium is a collaborative hub comprising around 90 entities dedicated to advancing life sciences through innovation.
The essence of this breakthrough lies in its potential to drastically expedite the drug discovery process. By accurately forecasting the shape and movement of proteins targeted by drugs, researchers can significantly reduce the time required for preliminary stages of development. For instance, a task that previously occupied specialists for an entire day can now be completed in a mere two hours.
Traditional methods of protein analysis involve labor-intensive scrutiny of numerous electron microscope images to infer a protein’s three-dimensional structure. Furthermore, predicting the precise movements of these proteins has presented substantial challenges, often extending the research timeline to several months.
Fujitsu and Riken’s ambition extends to creating an AI capable of predicting protein movements at the atomic level, drawing on genetic information. This advancement is expected to find applications in drug discovery within the next five years, according to Takashi Kato from Fujitsu’s Artificial Intelligence Laboratory.
Integrating generative AI into the initial phases of drug development allows for a more focused allocation of resources toward viable drug candidates, thereby enhancing the likelihood of successful outcomes. This approach not only promises to streamline the development of treatments for infectious diseases but also projects a substantial economic impact. The McKinsey Global Institute estimates that the application of generative AI in pharmaceuticals and medical products could generate an annual economic benefit ranging from $60 billion to $110 billion, encompassing savings on development costs and revenue from new drugs.
