Google Unveils AlphaFold 3 AI Model That Predicts Behavior of Biomolecules

Google DeepMind and Isomorphic Labs have jointly unveiled a more powerful version of AlphaFold, an artificial intelligence technology that helps scientists understand the behavior of the microscopic mechanisms that drive the cells in the human body.

An early iteration of AlphaFold, launched in 2020, successfully tackled the long-standing challenge known as the "protein folding problem," which had confounded scientists for over five decades.

Traditionally, biologists invested significant time and effort in determining the precise structure of individual proteins. However, AlphaFold revolutionized this process by swiftly predicting the three-dimensional configuration of a protein when given its amino acid sequence.

After its public release a year later, AlphaFold was used to expedite drug discovery efforts. Biologists swiftly adopted the technology to aid in understanding the Coronavirus and preparing for future pandemics. Additionally, researchers utilized AlphaFold to address pressing medical challenges such as malaria and Parkinson's disease.

The latest version of AlphaFold-AlphaFold3- can predict the behavior of other microscopic biological mechanisms, including DNA and RNA, in addition to predicting protein shapes.

The company also has a dedicated website where scientists can use AlphaFold3. Reportedly, other labs, most notably one at the University of Washington, offer similar technology.

“Biology is a dynamic system. You need to understand the interactions between different molecules and structures,” said Demis Hassabis, Google DeepMind’s chief executive and the founder of Isomorphic Labs, which Google also owns. “This is a step in that direction.”