Protein Structure · DeepMind / EMBL-EBI
AlphaFold DB Lookup
Retrieve pre-computed AlphaFold protein structures and confidence scores across 247 million UniProt entries.
What AlphaFold DB Lookup is
AlphaFold DB Lookup provides direct access to the AlphaFold Protein Structure Database developed by DeepMind and EMBL-EBI. The resource contains 247 million predicted protein structures covering organisms across the tree of life.
Instead of running computationally intensive de novo structure predictions, researchers can query an existing UniProt accession to obtain three-dimensional coordinates immediately. The tool also provides per-residue confidence scores and metadata, helping determine whether an existing entry is sufficient or if a fresh prediction is warranted.
Key facts
- Version
- 4.0
- Database size
- 247 million structures
- Hardware requirement
- CPU
- Domain
- Protein structure
What you can use it for
- Protein structure retrieval
- Obtain predicted all-atom coordinate files in PDB format for known proteins using their UniProt accession.
- Structural confidence assessment
- Evaluate per-residue pLDDT scores on a scale of 0 to 100 to determine which regions of a protein meet the reliability threshold of 70.
- Pre-computation screening
- Confirm whether a protein already has a pre-calculated model available before allocating graphics processing unit resources for new structural predictions.
Run it here
No installation and no GPU of your own: fill in the inputs and it runs on our servers. Results stay in your browser unless you sign in.
Run this model
AlphaFold DB Lookup
Strengths and limits
Good at
- Provides immediate access to 247 million pre-computed structures without requiring GPU hardware.
- Delivers per-residue confidence metrics alongside full atomic coordinates.
- Supplies supplementary metadata including organism, gene name, sequence length, and model version.
Watch out for
- Requires a valid UniProt accession identifier to perform a search.
- Does not compute novel structural predictions for uncatalogued sequences directly.
Inputs and outputs
You provide
UniProt accession
UniProtKB accession, e.g. P00533 (human EGFR).
You get back
All-atom coordinates in PDB format, viewable and downloadable.
pLDDT score per residue (0–100). Above 70 is generally reliable.
Organism, gene, sequence length and model version.
Licence
Commercial use permitted · CC BY 4.0
The CC BY 4.0 licence allows use in commercial research and products.
Questions
Can I use AlphaFold DB Lookup commercially?
Commercial use permitted. The CC BY 4.0 licence allows use in commercial research and products.
How do I run AlphaFold DB Lookup without installing anything?
Use the form under “Run it here” on this page: fill in the inputs and press Run model. It runs on our servers, typically in 700ms, and a run costs free. Visitors without an account get 1,000 credits a month.
What inputs are required to retrieve a structure?
The query requires a standard UniProt accession identifier.
What outputs are provided by the lookup?
It returns all-atom coordinates in PDB format, a vector containing per-residue pLDDT confidence scores, and a metadata table containing the organism, gene, sequence length, and model version.
How are the prediction confidence scores interpreted?
The system reports a pLDDT score for each residue ranging from 0 to 100, where values above 70 are generally considered reliable.
Does running this lookup require dedicated GPU resources?
No, structure retrieval runs entirely on standard CPU hardware.
Written from AlphaFold DB Lookup's own documentation. Statements citing a figure that does not appear there are removed before publication. Licence and running details come from our registry.