Bipartite Heterogeneous Network Method Based on Co-neighbor for MiRNA-Disease Association Prediction.

Min Chen,Yi Zhang,Ang Li,Wenhua Liu,Zheng Chen,Zejun Li

doi:10.3389/fgene.2019.00385

Abstract

In recent years, miRNA variation and dysregulation have been found to be closely related to human tumors, and identifying miRNA-disease associations is helpful for understanding the mechanisms of disease or tumor development and is greatly significant for the prognosis, diagnosis, and treatment of human diseases. This article proposes a Bipartite Heterogeneous network link prediction method based on co-neighbor to predict miRNA-disease association (BHCN). According to the structural characteristics of the bipartite network, the concept of bipartite network co-neighbors is proposed, and the co-neighbors were used to represent the probability of association between disease and miRNA. To predict the isolated diseases and the new miRNA based on the association probability expressed by co-neighbors, we utilized the similarity between disease nodes and the similarity between miRNA nodes in heterogeneous networks to represent the association probability between disease and miRNA. The model's predictive performance was evaluated by the leave-one-out cross validation (LOOCV) on different datasets. The AUC value of BHCN on the gold benchmark dataset was 0.7973, and the AUC obtained on the prediction dataset was 0.9349, which was better than that of the classic global algorithm. In this case study, we conducted predictive studies on breast neoplasms and colon neoplasms. Most of the top 50 predicted results were confirmed by three databases, namely, HMDD, miR2disease, and dbDEMC, with accuracy rates of 96 and 82%. In addition, BHCN can be used for predicting isolated diseases (without any known associated diseases) and new miRNAs (without any known associated miRNAs). In the isolated disease case study, the top 50 of breast neoplasm and colon neoplasm potentials associated with miRNAs predicted an accuracy of 100 and 96%, respectively, thereby demonstrating the favorable predictive power of BHCN for potentially relevant miRNAs.

Highlights

MiRNAs are a class of noncoding RNA molecules that play important roles in various biological processes, including proliferation, differentiation, aging, development, and apophasis (Ambros, 2004)
We proposed a link prediction method of bipartite heterogeneous network based on co-neighbors to predict miRNA-disease association (BHCN)
We tested the model predictive performance with eight similarity indexes in six different scenarios which are as follows:predictive performance using only known miRNA-disease association information (BHCNMDA); predictive performance based on miRNA similarity without miRNA similarity network reconstruction (BHCN-MSnoMSR); predictive performance based on miRNA similarity with miRNA similarity network reconstruction using miRNA family information (BHCN-MS-MSR); predictive performance based on disease similarity without disease similarity network reconstruction (BHCN-DS-noDSR); predictive performance based on disease similarity with disease similarity network reconstruction using known miRNA-disease association information (BHCN-DS-DSR) and predictive performance using all information (BHCN)

Summary

Introduction

MiRNAs are a class of noncoding RNA molecules that play important roles in various biological processes, including proliferation, differentiation, aging, development, and apophasis (Ambros, 2004). Further exploration of the relationship between miRNAs and diseases can help elucidate the pathogenesis of diseases. Traditional experimental methods such as PCR and microarray (Chen et al, 2009) can reveal the relationship between miRNA and disease, but time consuming and only applicable to small-scale experimental data. In the past few years, many computational methods that predict the association between miRNA and diseases were suggested to find the association between miRNA and disease on a large scale (Alaimo et al, 2014; Zou et al, 2015b; Chen et al, 2017f; Chen and Qu, 2018)

Methods

Results

Conclusion