Google AI predicts the structure and interactions of all molecules in the biological world in HD

Within each cell, there are billions of molecular machines that are crucial to understanding and treating diseases. The latest version of AlphaFold, an artificial intelligence system developed by Google, can predict the structure and interactions of all molecules of life. This advancement has been described as bringing “the biological world to high definition” and allows scientists to see cellular systems in all their complexity, including structures, interactions, and modifications.

DeepMind, in collaboration with Isomorphic Labs, developed this revolutionary model that builds upon the success of AlphaFold 2. This new model can accurately predict the structure of a wide range of biomolecular systems and improve protein interactions with other molecules by at least 50% compared to existing methods. This innovative AI model has the potential to transform various areas of science, from developing biorenewable materials to accelerating drug design and genomic research.

AlphaFold 3 is capable of generating joint three-dimensional structures of a list of molecules, illustrating how they fit together. It can model large biomolecules like proteins, DNA, and RNA, as well as small molecules and chemical modifications that control cell function and can cause disease when altered. By providing a precise vision at the atomic level of biomolecular systems, this AI model helps researchers understand how complex networks of molecules interact in cells, which is crucial for treating diseases through rational drug design.

Scientists can access most features of AlphaFold 3 for free through the AlphaFold server, enabling them to model structures and ask bold questions to accelerate discoveries. This new tool offers a deeper understanding of the biomolecular world within us and how molecules interact in cells, ultimately contributing to the development of new treatments and advancements in various scientific fields.