AlphaFold
#6google-deepmind · seit 2018-12-01 · 10× · zuletzt 30. Juni 2026
AlphaFold is an AI system by Google DeepMind (co-developed with Isomorphic Labs for version 3) that predicts 3D structures of proteins and other biomolecules from amino acid sequences. The current version, AlphaFold 3 (released May 2024), extends predictions to proteins, DNA, RNA, ligands, and ions, including their molecular interactions. Scientists can access it for free via the AlphaFold Server for non-commercial research; the database of over 200 million protein structures (alphafold.ebi.ac.uk, in partnership with EMBL-EBI) is freely available under CC-BY-4.0. AlphaFold 2 was awarded the 2024 Nobel Prize in Chemistry and is used by over 3 million researchers in more than 190 countries.
Fonctionnalités
| Deployment Model | Cloud/web (AlphaFold Server, AlphaFold DB) as well as local/HPC installation via open-source code and model weights (e.g., supercomputing clusters) |
| Use Case Scope | Prediction of protein 3D structures and interactions with DNA, RNA, ligands/ions for biological and biomedical research (incl. drug research, disease research) |
| Integrations | Google Colab (official notebook), ColabFold (MMseqs2-based), UniProt/PDB linkage in AlphaFold DB, EMBL-EBI 3D-Beacons |
| License | AlphaFold 3 source code: Apache 2.0; model parameters: proprietary, non-commercial Terms of Use; AlphaFold 2 parameters: CC BY-NC 4.0; AlphaFold DB data: CC-BY-4.0 |
| Platform | Web service (AlphaFold Server), database (AlphaFold DB), open-source code (GitHub), Google Colab notebook |
| Price | AlphaFold DB & AlphaFold Server: free (non-commercial use); AlphaFold code/parameters: free under respective license |
| Release Date | AlphaFold 3 announced on May 8, 2024 |
