AlphaFold 3 Can Now Predict the Structure of All Life's Molecules, Accelerating Drug Discovery

Google DeepMind and Isomorphic Labs unveiled AlphaFold 3 in May 2024, a major evolution of the AI system that revolutionized structural biology. While its predecessor AlphaFold 2 could predict the 3D structure of individual proteins from their amino acid sequences, AlphaFold 3 goes dramatically further: it can predict the structure and interactions of virtually all biological molecules, including proteins, DNA, RNA, ligands, and small molecules used in drug development.

Published in the journal Nature, the research demonstrated that AlphaFold 3 achieves at least 50% better accuracy than existing methods for predicting how proteins interact with other molecules. For some categories, such as protein-ligand interactions crucial for drug design, the improvement is even greater. The system uses a novel diffusion-based architecture — similar in concept to the technology behind AI image generators — that assembles molecular structures atom by atom.

“While its predecessor AlphaFold 2 could predict the 3D structure of individual proteins from their amino acid sequences, AlphaFold 3 goes dramatically further: it can predict the structure and interactions of virtually all biological molecules, including proteins, DNA, RNA, ligands, and small molecules used in drug development.”

The implications for drug discovery are transformative. Traditionally, determining how a drug candidate molecule binds to its target protein requires months of expensive laboratory work, including X-ray crystallography or cryo-electron microscopy. AlphaFold 3 can generate these predictions computationally in minutes, allowing researchers to rapidly screen and optimize potential drug molecules before they ever enter a laboratory.

Isomorphic Labs, a DeepMind spin-off focused on drug discovery, has already signed partnerships worth over $3 billion with major pharmaceutical companies Eli Lilly and Novartis to apply AlphaFold 3 to their drug development pipelines. The company is using the system to identify and design new drug candidates for diseases that currently lack effective treatments.

DeepMind made the AlphaFold Server freely available to academic researchers worldwide, allowing scientists at universities and non-profit organizations to use the tool for their own research at no cost. Within weeks of launch, the server had processed millions of prediction requests from researchers across over 100 countries.

The achievement builds on the legacy of AlphaFold 2, which has already been used by over 1.8 million researchers since its public release in 2021 and has been credited with contributing to advances in malaria vaccines, cancer research, plastic-degrading enzymes, and antibiotic development.