HSPred Help

Interface definition

The user needs to specify the chains forming the interface by entering the chain identifiers for protein 1 and protein 2. The chains specified must be present in the submitted PDB file.


HSPred makes a prediction for each amino acid at the interface (excluding Proline and Glycine residues). It returns a 3 columns table which reports (i) a residue identifier (in the form of chain identifier plus sequence number), (ii) the residue identity (single letter code) and (iii) HSPred output score. A score greater than zero represents a predicted hot spots, i.e. a residue that if mutated to alanine leads to a change in free energy of binding greater than 2 kcal/mol. Negative scores are predicted non hot spots.  

The results are visualised using Jmol. The structures of the two interacting proteins are displayed in separate windows.  Predictions are mapped onto the temperature factor column of the uploaded PDB file. Predicted hot spots residues are coloured red and non hot spot residues are coloured white. Amino acids that are not part of the interface (or Pro and Gly at the interface) are in blue. It is also possible to download these prediction PDB files, in order to visualize them using other molecular graphics viewing programs.