Downloads
The AlphaFold DB website currently provides bulk downloads for the 48 organisms listed below, as well as the majority of Swiss-Prot.
You can download a prediction for an individual model by visiting the corresponding structure page
(examples: https://alphafold.ebi.ac.uk/entry/AF-F4HVG8-2-F1, https://alphafold.ebi.ac.uk/entry/AF-0000000078834380).
For downloading all predictions for a given species, use the download links below. Note that this option is only available on the desktop version of the site.
The uncompressed archive files (.tar) contain all the available compressed PDB and mmCIF files (.gz) for a reference proteome. In the case of proteins longer than 2700 amino acids (aa), AlphaFold provides 1400aa long, overlapping fragments. For example, Titin has predicted fragment structures named as Q8WZ42-F1 (residues 1–1400), Q8WZ42-F2 (residues 201–1600), etc. These fragments are currently only available for the human proteome in these proteome archive files, not on the website.
For downloading predictions for all the species in the tables below, visit the FTP site: https://ftp.ebi.ac.uk/pub/databases/alphafold
Compressed prediction files for model organism proteomes:
| Species | Common Name | Reference Proteome | Predicted Structures | Download |
|---|
Compressed prediction files for global health proteomes:
| Species | Common Name | Reference Proteome | Predicted Structures | Download |
|---|
Compressed prediction files for Swiss-Prot:
| File type | Predicted Structures | Download |
|---|
Collaborators dataset
Selected collaborator datasets are distributed via the FTP area. Coordinate files are provided in chunked archives in the collaborations subdirectory. Where available, corresponding MSAs are provided in the msas subdirectory. Dataset-specific availability may vary.
For further details about individual datasets, please consult the corresponding dataset FAQ.
Licence and attribution
Data is available for academic and commercial use, under a CC-BY-4.0 licence.
If you make use of an AlphaFold prediction, please cite the following papers:
Fleming J. et al. AlphaFold Protein Structure Database and 3D-Beacons: New Data and Capabilities. Journal of Molecular Biology, (2025)
And the relevant structure’s publication, please see the entry page, dataset collection or structure metadata for details.
Please also cite the following publications for predictions from:
- UniProt dataset: Jumper, J et al. Highly accurate protein structure prediction with AlphaFold. Nature (2021)
- Big Fantastic Virus Database: Kim, RS et al. BFVD—a large repository of predicted viral protein structures. NAR (2024)
- AllTheBacteria dataset: Hunt M, Lima L, Anderson D, Bouras G, Hall M, Hawkey J, Schwengers O, Shen W, Lees JA, Zamin Iqbal Z. BioRXiV (2025)
- Wheeler Lab dataset: Wheeler RJ. A resource for improved predictions of Trypanosoma and Leishmania protein three-dimensional structure. PLoS One (2021)
- Viro3D dataset (2025): Litvin, U et al. Viro3D: a comprehensive database of virus protein structure predictions. Mol Syst Biol (2025)
- VAD dataset (2026): Odai, R., Leemann, M., Al-Murad, T., Abdullah, M., Shyrokova, L., Tenson, T., Hauryliuk, V., Durairaj, J., Pereira, J., Atkinson, G.C. The Viral AlphaFold Database of monomers and homodimers reveals conserved protein folds in viruses of bacteria, archaea, and eukaryotes. Science Advances (2025)
If you use data from AlphaMissense in your work, please cite the following paper:
AlphaMissense Copyright (2023) DeepMind Technologies Limited.
AlphaFold Data provided by Google DeepMind:
AlphaFold Data Copyright (2022) DeepMind Technologies Limited.
For AlphaFold Data provided by 3P data providers:
Kinetoplastid data Copyright (2025) Wheeler Lab
AllTheBacteria data Copyright (2025) AllTheBacteria Consortium
BFVD Data Copyright (2025) The BFVD Development Team
Viro3D data (2025)
Viral AlphaFold Database Copyright (2025) Lund University
Feedback and questions
If you want to share feedback on an AlphaFold structure prediction, consider using the feedback buttons at the bottom of each structure viewer. If you have any questions that are not covered in the FAQs, please contact alphafold@deepmind.com. If you have feedback on the website or experience any bugs, please contact afdbhelp@ebi.ac.uk.
Disclaimer
All AlphaFold and AlphaMissense Data and other information provided on this site contain predictions with varying levels of confidence, is for theoretical modelling only and caution should be exercised in its use. It is provided 'as-is' without any warranty of any kind, whether expressed or implied. For clarity, no warranty is given that use of the information shall not infringe the rights of any third party. The information is not intended to be a substitute for professional medical advice, diagnosis, or treatment, and does not constitute medical or other professional advice. The AlphaFold and AlphaMissense Data have not been validated for, and are not approved for, any clinical use.
Use of the AlphaFold Protein Structure Database is subject to EMBL-EBI Terms of Use.
