Prediction of protein-protein interaction types using the decision templates

Protein-protein interactions (PPIs) play a key role in many cellular processes. Knowing about the multitude of PPIs can help the biological scientist understand the molecular mechanism of the cell. Unfortunately, it is both time-consuming and expensive to do so solely based on experiments. Therefore, developing computational approaches for predicting PPIs, PPI binding sites and PPI types would be of significant value. Here, we propose a novel method for predicting the PPI types based on decision templates. First, we introduce the concept of tensor product to construct three kinds of feature vectors which are the amino acid composition tensor product, the residue multi-scale conservation energy tensor product and the secondary structure content tensor product. Then, the correlation-based feature selection method was also used to reduce the dimensionality of these feature vectors. So, the protein pair can be represented by our three new kinds of feature vectors and Zhu's six kinds of feature vectors. The nine kinds of feature vectors are further taken as the inputs of individual support vector machine classifier respectively, and the outputs of these classifiers are aggregated with decision templates in decision level. The overall success rate obtained by jackknife cross-validation was 90.95%, indicating our method is very promising for predicting PPI types.