AlphaFold 3 predicts joint structures of proteins, nucleic acids, ligands and ions
What was demonstrated
Google DeepMind and Isomorphic Labs described a diffusion-based model that predicts the joint 3D structure of complexes containing proteins, nucleic acids, small molecules, ions and modified residues. On the authors' benchmarks it was more accurate than state-of-the-art docking tools for protein-ligand interactions, than nucleic-acid-specific predictors for protein-nucleic acid interactions, and than AlphaFold-Multimer v2.3 for antibody-antigen prediction.
Why it matters
A single model covering most biomolecular interaction types reduces the need for separate specialist tools and could speed structure-based drug discovery and protein design.
Caveats
- Accuracy figures come from computational benchmarks against known structures, not prospective experimental validation of new predictions.
- Predicted structures are static and do not capture conformational dynamics.
- At publication, code and weights were not released for unrestricted use, which limited independent reproduction; access arrangements have since changed and should be checked.
Technology profiles
Sources
- peer reviewed paper
Accurate structure prediction of biomolecular interactions with AlphaFold 3
Josh Abramson, Jonas Adler, Jack Dunger et al.. Nature, 8 May 2024. DOI: 10.1038/s41586-024-07487-w.
Last reviewed 20 Oct 2026.