Graph Kernels and Applications in Protein Classification

Abstract

Protein classification is a well established research field concerned with the discovery of molecule’s properties through informa­tional techniques. Graph-based kernels provide a nice framework combining machine learning techniques with graph theory. In this paper we introduce a novel graph kernel method for an­notating functional residues in protein structures. A structure­ is first modeled as a protein contact graph, where nodes ­corres­pond to residues and edges connect spatially neighboring resi­dues. In experiments on classification of graph models of proteins, the method based on Weisfeiler-Lehman shortest path kernel with complement graphs outperformed other state-of-art methods.