The tool PISA is designed to do this

http://www.ebi.ac.uk/msd-srv/prot_int/pistart.html

However, you should always use experimental methods to verify what computations indicate.

Thank You Christina. It is really helpful for me.

@ Ahmed. Thanks for your link but I was looking for a 3D tool that was suggested by Christina.

FoldX will be an another option.

Module to use:

FoldX says: "The program just mutates every residue on the interface of the selected molecules to Ala and produces the ΔΔG of the mutation."

In addition to Priybrata Panigrahi, I would advice you to consider computationnal alanine scanning.

An alternative of FoldX based on molecular dynamics simulation and MM/PBSA estimation of the binding free energy using AmberTools. ( http://ambermd.org/ ; http://ambermd.org/tutorials/advanced/tutorial3/py_script/section3.htm )

It has been succesfully applied to tubulin to identify identify the amino acids responsible for tubulin–tubulin binding and thus to design anti-mitotic peptides.

Pieraccini S, Saladino G, Cappelletti G, Cartelli D, Francescato P, Speranza

G, Manitto P, Sironi M. In silico design of tubulin-targeted antimitotic

peptides. Nat Chem. 2009 Nov;1(8):642-8. doi: 10.1038/nchem.401. Epub 2009 Oct

23. PubMed PMID: 21378956.

http://www.nature.com/nchem/journal/v1/n8/abs/nchem.401.html

http://kfc.mitchell-lab.org/

KFC2 Server is a server for prediction of hot-spot interface residues.