Predicting Candidate Genes Based on Combined Network Topological Features: A Case Study in Coronary Artery Disease

Liangcai Zhang,Xu Li,Jingxie Tai,Wan Li,Lina Chen,Weili Zhang

doi:10.1371/journal.pone.0039542

Abstract

Predicting candidate genes using gene expression profiles and unbiased protein-protein interactions (PPI) contributes a lot in deciphering the pathogenesis of complex diseases. Recent studies showed that there are significant disparities in network topological features between non-disease and disease genes in protein-protein interaction settings. Integrated methods could consider their characteristics comprehensively in a biological network. In this study, we introduce a novel computational method, based on combined network topological features, to construct a combined classifier and then use it to predict candidate genes for coronary artery diseases (CAD). As a result, 276 novel candidate genes were predicted and were found to share similar functions to known disease genes. The majority of the candidate genes were cross-validated by other three methods. Our method will be useful in the search for candidate genes of other diseases.

Highlights

Many complex diseases like coronary artery disease result from a complex interplay of multiple genes
Previous discoveries [4,5,6,7,8] demonstrated that direct interaction partners of a protein are likely to share similar functions with it, and causative genes of some complex disease tends to reside in the same network communities such as biological modules, protein complexes, pathways or subnetworks of a given biological network
We found that a majority of coronary artery diseases (CAD) candidate genes that had similar network topological features tended to have a significantly functional relatedness to known disease genes in following categories such as protein binding, receptor binding, molecular transducer activity, signal transducer activity, receptor activity, oxidoreductase activity, hydroxymethylglutaryl-CoA reductase (NADPH) activity and so on

Summary

Introduction

Many complex diseases like coronary artery disease result from a complex interplay of multiple genes. Recent accumulation of reliable molecular interaction data has boosted progress in the discovery of novel susceptibility genes and fueled expectations about opportunities of computational approaches for distinguishing disease-related genes from non-disease ones. Previous discoveries [4,5,6,7,8] demonstrated that direct interaction partners of a protein are likely to share similar functions with it, and causative genes of some complex disease tends to reside in the same network communities such as biological modules, protein complexes, pathways or subnetworks of a given biological network. Apart from that, with the accumulation of human proteinprotein interaction networks, it is necessary to introduce novel approaches to find out effective network topological features for gene classifications, and further aid in the prediction of candidate disease genes

Methods

Results

Conclusion