A Novel Network-Based Computational Model for Prediction of Potential LncRNA⁻Disease Association.

Yang Liu,Zhanwei Xuan,Xiang Feng,Lei Wang,Haochen Zhao

doi:10.3390/ijms20071549

Abstract

Accumulating studies have shown that long non-coding RNAs (lncRNAs) are involved in many biological processes and play important roles in a variety of complex human diseases. Developing effective computational models to identify potential relationships between lncRNAs and diseases can not only help us understand disease mechanisms at the lncRNA molecular level, but also promote the diagnosis, treatment, prognosis, and prevention of human diseases. For this paper, a network-based model called NBLDA was proposed to discover potential lncRNA–disease associations, in which two novel lncRNA–disease weighted networks were constructed. They were first based on known lncRNA–disease associations and topological similarity of the lncRNA–disease association network, and then an lncRNA–lncRNA weighted matrix and a disease–disease weighted matrix were obtained based on a resource allocation strategy of unequal allocation and unbiased consistence. Finally, a label propagation algorithm was applied to predict associated lncRNAs for the investigated diseases. Moreover, in order to estimate the prediction performance of NBLDA, the framework of leave-one-out cross validation (LOOCV) was implemented on NBLDA, and simulation results showed that NBLDA can achieve reliable areas under the ROC curve (AUCs) of 0.8846, 0.8273, and 0.8075 in three known lncRNA–disease association datasets downloaded from the lncRNADisease database, respectively. Furthermore, in case studies of lung cancer, leukemia, and colorectal cancer, simulation results demonstrated that NBLDA can be a powerful tool for identifying potential lncRNA–disease associations as well.

Highlights

In recent years, accumulating evidence studies have shown that non-coding RNAs are involved in various biological processes in the human body [1,2,3], and long non-coding RNAs, as a class of important heterologous ncRNAs with a length greater than 200 nt, play critical roles in various human biological processes such as chromatin modification, cell differentiation, proliferation and apoptosis, translational and post-translational regulation, and so on [4,5,6]
Detecting potential long non-coding RNAs (lncRNAs)–disease associations can help us understand the pathogenesis of human diseases at the molecular level, and further facilitate the diagnosis, treatment, and prevention of human diseases [10]
In contrast to the above machine learning-based models, according to the assumption that functionally similar lncRNAs show similar interaction patterns with similar diseases, Sun et al [16] proposed a computational model, RWRlncD, in which a global network was constructed first based on disease similarity, lncRNA functional similarity, and known lncRNA–disease associations, and a random walk with restart method was implemented on the newly constructed global network to infer potential lncRNA–disease associations

Summary

Introduction

In recent years, accumulating evidence studies have shown that non-coding RNAs (ncRNAs) are involved in various biological processes in the human body [1,2,3], and long non-coding RNAs (lncRNAs), as a class of important heterologous ncRNAs with a length greater than 200 nt, play critical roles in various human biological processes such as chromatin modification, cell differentiation, proliferation and apoptosis, translational and post-translational regulation, and so on [4,5,6]. The number of known lncRNA–disease associations in these databases is far from meeting the needs of modern medical researches, due to traditional biological experiment methods for discovering potential relationships between lncRNAs and diseases that are very expensive and time-consuming [13]. More and more researchers have devoted efforts to constructing computational models to identify potential relationships between lncRNAs and diseases. In contrast to the above machine learning-based models, according to the assumption that functionally similar lncRNAs show similar interaction patterns with similar diseases, Sun et al [16] proposed a computational model, RWRlncD, in which a global network was constructed first based on disease similarity, lncRNA functional similarity, and known lncRNA–disease associations, and a random walk with restart method was implemented on the newly constructed global network to infer potential lncRNA–disease associations. Zhao et al [19] developed a distance correlation set-based computational model, DCSMDA, to predict potential miRNA–disease associations, in which a tripartite miRNA–lncRNA–disease network was constructed through integrating disease similarity, miRNA similarity, and lncRNA similarity

Methods

Results

Conclusion