P2Rank

Find potential ligand-binding pockets in a protein structure and rank the candidate sites.

API/CLI tool ID: p2rank

Understanding inputs and outputs

A pocket is a region on a protein surface that may accommodate or contact another molecule. P2Rank identifies candidate regions in the supplied protein structure. This differs from docking, which uses a ligand structure to calculate binding poses.

score and probability describe how promising the model considers a region as a pocket. A probability of 0.8 does not mean that a particular molecule has an 80% chance of binding. No candidates does not necessarily mean the run failed; the model may simply have found no reportable pockets in the input structure.

Input settings

SettingFormatRequiredDefaultDescription and constraints
proteinPdbpdbRequired—Target protein structure — Extensions pdb · cif

Output metrics

ColumnDisplay nameRangeDescription
probabilityPocket probability0~1Probability, from the trained model, that this site is a ligand pocket
scorePocket score—Ligandability score: higher values suggest a site more likely to bind a ligand. Not evidence of real binding or efficacy.

See Reading results for metric definitions and interpretation. A range shows possible values; whether higher or lower is better depends on the metric.

Execution availability

Open an execution plan in your project to see available tool versions and costs. Review and approve the inputs, settings, and terms before running.

License

Code, model weights, and accompanying data have their own terms. The table summarizes terms from official sources; check your execution plan for the configuration you will use. License texts are available in Third-party notices.

ComponentLicenseCommercial useSourceNotes
CodeMITAllowed, subject to notice requirementsLicense textTrained model files included in the official package are distributed under the same repository terms
Model weightsMITAllowed, subject to notice requirementsLicense textThe trained model (models/default/model.zst) has no separate license and is distributed with the same MIT text at the distribution root. The LICENSE.txt inside the 2.5.1 release asset (sha256 d243f2d9…) is byte-identical (checked 2026-11-09)

Pocket detection and ligandability ranking (single Java distribution). License text read directly on 2026-10-26; model license text added on 2026-11-09

Checked on 2026-11-09.

References

On this page