CNetA: Network alignment by combining biological and topological features

Abstract

Due to the rapid progress of high-throughput techniques in past decade, a lot of biomolecular networks are constructed and collected in various databases. However, the biological functional annotations to networks do not keep up with the pace. Network alignment is a fundamental and important bioinformatics approach for predicting functional annotations and discovering conserved functional modules. Although many methods were developed to address the network alignment problem, it is not solved satisfactorily. In this paper, we propose a novel network alignment method called CNetA, which is based on the conditional random field model. The new method is compared with other four methods on three real protein-protein interaction (PPI) network pairs by using four structural and five biological criteria. Compared with structure-dominated methods, larger biological conserved subnetworks are found, while compared with the node-dominated methods, larger connected subnetworks are found. In a word, CNetA preferably balances the biological and topological similarities.