Binding site extraction by similar subgraphs mining from protein molecular surfaces

Abstract

Most proteins express their functions by binding with other proteins or molecular compounds called ligands. The local portion involved in binding is called a binding site. The characteristics of the binding site often determine the function of the protein, so clarifying the location of the binding site of the protein helps analyze the function of proteins. Binding sites that bind to similar ligands often have common surface structures. Such common structures are called surface motifs. Therefore, extracting the surface motifs among several proteins with similar functions improves binding site prediction. We propose a method of predicting binding sites by extracting the surface motifs that are frequently observed in only a specific group, which means a set of proteins that bind to the same ligand. Since most binding sites have concave structures called pockets, the pockets are compared and common structures are searched for to extract the surface motifs by applying similar graph mining to the pocket data, which are represented as graphs, to find the frequent subgraphs among the pockets of several proteins. In addition, the common binding sites across several groups can be predicted in such a way to integrate more than one group. Applying our proposed method to a set of 37 proteins of five groups, we achieved success rates of binding site prediction over 40% and 50% for more than half of the groups without group integration and using integration, respectively.