Description
Researchers have performed a systematic analysis of protein-ligand and protein-protein complexes in the Protein Data Bank (PDB) to identify clusters of interface residues that delineate the binding site of a high affinity small molecule at the PPI interface. These clusters are referred to as Small Molecule Inhibitor Starting Points (SMISPs), since the binding mode of these residues in the acceptor protein provides a starting point for small molecule inhibitor design. The SMISPs form the basis of a training set where each cluster is characterized using a variety of properties commonly used to predict hot spots in PPIs, e.g., free energies, solvent accessible surface area, and sequence conservation. The training set is used to train two distinct classifiers, a support vector machine and an easily interpreted rule classifier. Both classifiers achieve better than 70% leave-one-complex-out cross validation accuracy and correctly predict SMISPs of known PPI inhibitors not in the training set. Application to non-redundant PPIs in the PDB predicts that 48% possess a sufficient quantity and quality of SMISPs to be susceptible to small molecule intervention.IP Status
Research Tool
