Automated Hub-Protein Detection via a New Fused Similarity Measure-Based Multi-objective Clustering Framework

Abstract

In the field of computational biology and bioinformatics, there have been limited studies on the development of protein-protein proximity measures which blend multiple sources of biological properties of protein. In Protein-Protein Interaction Network (PPIN), hub-proteins play a central role. There are many literature with user-studied different degree cut-offs for defining hub-proteins. Therefore, there is a need for a standard method for identifying hub-proteins without manually determining the degree cut-off. In the current research article, an effort has been made towards addressing both problems. At first, we have proposed a new Fused protein-protein Similarity measure - FuSim, which involves biological properties of both Gene Ontology (GO) and PPIN. Later, utilizing the proposed similarity measure, a multi-objective clustering algorithm-based automated hub-protein detection framework is developed.