Centrality Measure Analysis on Protein Interaction Networks

Abstract

Analysis of protein interaction is widely recognized to understand cell physiology and disease conditions. The increase in the accumulation of these interaction data facilitates the recognition of the essential proteins in Protein Protein Interaction (PPI) networks. An array of centrality measures are available to uncover essential proteins in PPI networks. However, majority approaches are centered around topological properties of PPI. Few approaches integrate gene annotation with topology for predicting essential proteins. This biological framework in PPI network are inferred in terms of graph-theoretic approaches. The topological analysis focuses on protein, their interactions, and the subnetworks. In this research, we review the common centrality measures. We thoroughly studied the centrality aspect of each node in the PPI to detect the influential nodes and the impact of topological features in centrality measures. We applied centrality measures to the PPI networks obtained from the Biological General Repository for Interaction Networks (BioGRID) and Mammalian Protein Protein Database (MIPS) datasets. The experimental evaluation shows the behavior of centrality measures to the datasets.