The KFC Server (Knowledge-based FADE and Contacts) predicts binding "hot spots" within protein-protein interfaces by recognizing structural features indicative of important binding contacts. The server analyzes several chemical and physical features surrounding an interface residue and predicts the classification of the residue using a model trained on prior experimental data.

The KFC2 Server manual can be found here.

If you use the KFC2 Server, please cite:
        S. J. Darnell, D. Page, and J. C. Mitchell
        Automated Decision-Tree Approach to Predicting Protein-Protein Interaction Hot Spots.
        Proteins, 68(4): 813-823, 2007.
        View Abstract
and:
        X. Zhu, and J. C. Mitchell
        KFC2: A knowledge-based hot spot prediction method based on interface solvation,
        atomic density and plasticity features.
        Proteins, 79(9): 2671-2683, 2011.
        View Abstract