Assessing the druggability of protein-protein interactions by a supervised machine-learning method

To assess the druggability of the PPIs, 69 attributes were selected to cover a wide range of structural, drug and chemical, and functional information on the PPIs. These attributes were used as feature vectors in the SVM-based method. Thirty PPIs known to be druggable were carefully selected from previous studies; these were used as positive instances. Our approach was applied to 1,295 human PPIs with tertiary structures of their protein complexes already solved. The best SVM model constructed discriminated the already-known target PPIs from others at an accuracy of 81% (sensitivity, 82%; specificity, 79%) in cross-validation. Among the attributes, the two with the greatest discriminative power in the best SVM model were the number of interacting proteins and the number of pathways.Using the model, we predicted several promising candidates for druggable PPIs, such as SMAD4/SKI. As more PPI data are accumulated in the near future, our method will have increased ability to accelerate the discovery of druggable PPIs.Interfering with PPIs by small ligands has been regarded as challenging mainly due to the flatness and large surface area of protein-protein interfaces [1]. However, targeting PPIs is a highly attractive strategy for therapeutic interventions, because most proteins function in cells by interacting with other proteins. To date, over 30 PPIs have been intensively studied as targets for PPI-inhibiting small ligands; these include MDM2/TP53, BCL-XL(BCL-2)/BAK, and IL2/IL2 receptor α [[1-7] and references therein]. The interfaces of these drug target PPIs are characterized by a concave, rather than flat, surface and so-called 'hot spots', which is a small area within the interface containing a few amino acids that contribute a large fraction of binding free energy of the interaction [1]. Some PPI-inhibiting small ligands have been proven to have high potency in both in vitro and in vivo assays on models of human diseases such as cancer [8,9]. These studies stron