Downloading files

Coordinate files

To perform further analysis or visualise a protein structure outside of the AlphaFold Database website, you need to download its atomic coordinate file.

A coordinate file is a text-based, digital representation of a molecule’s three-dimensional structure. It contains a list of all atoms in the protein, specifying their precise position in 3D space using a Cartesian coordinate system (X, Y, Z). 

For AlphaFold predictions, the per-residue confidence score (pLDDT) is stored in the B-factor column of the coordinate file. This structured data provides all the necessary information for molecular graphics programs to render an accurate and interactive 3D model.

You can download the coordinate files of a predicted structure in two standard formats.

• PDB file (.pdb): The legacy format for macromolecular structures. While widely supported, it has limitations regarding the size and complexity of the molecules it can represent.
• mmCIF file (.cif): The current standard format maintained by the Worldwide Protein Data Bank (wwPDB). The macromolecular Crystallographic Information File (mmCIF) is more robust and can accommodate much larger and more complex structures, making it the recommended choice.

These formats are widely accepted by 3D structure viewing software, such as PyMOL and ChimeraX.

The data files are versioned, and previous versions are available via FTP, but the web pages always display the latest version.

Predicted Aligned Error

The Predicted Aligned Error (PAE) is available for download as a .json file. This file is in a custom format and it isn’t supported by any existing software – you will have to use a programming language to analyse or plot the information that is contained in it.

The fields in the JSON file are:

• predicted_aligned_error: The PAE value of the residue pair, rounded to the closest integer. For the PAE value at position (i, j), i is the residue on which the structure is aligned, j is the residue on which the error is predicted.
• max_predicted_aligned_error: A number that denotes the maximum possible value of PAE. The smallest possible value of PAE is 0.

AlphaMissense data

For canonical human entries with AlphaMissense data, we provide downloadable files in csv format to help you explore the data:

• Heatmap data. This file contains the scores estimating the likelihood of pathogenicity and classifications for each possible amino acid substitution within the protein. This file is used to visualise the heatmap you see on the entry pages.
• Pathogenicity scores (HG19 and HG38). These files include AlphaMissense scores for all possible missense single-nucleotide variants across the human reference genome. Each file corresponds to a specific genome assembly with information on a specific genome position, where HG19 corresponds to the GRCh37 human genome reference assembly and HG38 to the GRCh38 reference assembly.

Multiple Sequence Alignment (MSA)

The multiple sequence alignments (MSAs) used to generate each prediction are now available for download as .a3m files; offering a rich new layer of data for evolutionary and structural analyses. These file can be inspected using your favourite alignment viewer, e.g. JalView.