AlloPred uses a novel method of normal mode analysis and pocket features to predict allosteric pockets on proteins. On a test set of known allosteric proteins the highest ranked pocket per protein was an allosteric pocket in 23 out of 40 cases. An allosteric pocket was ranked first or second in 28 out of 40 cases.

Job submission

Use example parameters View example results

Predict allosteric pockets on a protein. Either enter a PDB ID and chain(s) or upload a PDB file:

PDB ID and chain(s)
PDB file upload
Job name (optional)

Enter active site residues, e.g. for residue 10 on chain A and residue 11 on chain B enter 10:A,11:B.

Click here to attempt to extract active site residues from the Catalytic Site Atlas based on the PDB ID entered above.

Active site residues

Providing an email address means you will get an email when the job is complete.

User email (optional)

Job progress/results

View progress/results by entering the job key:

If you use AlloPred, please cite: Greener, JG and Sternberg, MJE, AlloPred: prediction of allosteric pockets on proteins using normal mode perturbation analysis, BMC Bioinformatics, 2015, 16(335). Link to paper.

Last updated 29/06/17