Research On The Prediction Model Of Disease-related Non-coding RNA Based On Random Walk

In recent years,Lnc RNAs(long-non-coding RNAs)have been proved to be closely related to the occurrence and development of many serious diseases that are seriously harmful to human health.However,most of the Lnc RNA-disease associations have not been found yet due to high costs and time complexity of traditional bio-experiments.Hence,it is quite urgent and necessary to establish efficient and reasonable computational models to predict potential associations between Lnc RNAs and diseases.In bioinformatics,using an effective prediction model to reveal the potential relationship between diseases and Lnc RNAs has become a hot research topic in recent years and has been highly concerned by many researchers.At the beginning of this paper,we first introduce the background knowledge of this research area and the research status both here and abroad.After that,we focus on two models both based on random walk to predict the potential associations of Lnc RNA-disease.The research contents of these two methods are briefly described as follows:(1)First,we obtained known Lnc RNA-disease associations data from the database and preprocessed it.(2)Then,we propose a new prediction model LRWHLDA based on local random walk to infer the potential relationship between human Lnc RNAs and diseases.In LRWHLDA,we first establish a new heterogeneous network by integrated the three associations of the disease-disease,Lnc RNA-Lnc RNA,and the known Lnc RNA-disease,which contains two types of nodes and three types of links.The setup of this heterogeneous network allows that LRWHLDA can be implemented in the case of lacking known Lnc RNA-disease associations.On this basis,a novel local random walk based prediction model called LRWHLDA is proposed for inferring potential associations between human Lnc RNAs and diseases,which can help LRWHLDA achieve high prediction accuracy but with low time complexity.this allows our prediction model LRWHLDA to remain efficient in a very dense and complex network.By means of a simulated experiment,it is easy to know that LRWHLDA contains the potential to be a representative of emerging methods in thefield of research on potential Lnc RNA-disease associations prediction.(3)Finally,a novel prediction model called TCSRWRLD is proposed to predict potential Lnc RNA-disease associations based on improved random walk with restart.In TCSRWRLD,a heterogeneous Lnc RNA-disease network is constructed first by combining the integrated similarity of Lnc RNAs and the integrated similarity of diseases.And then,for each Lnc RNA/disease node in the newly constructed heterogeneous Lnc RNA-disease network,it will establish a node set called TCS(Target Convergence Set)consisting of top 100 disease/Lnc RNA nodes with minimum average network distances to these disease/Lnc RNA nodes having known associations with itself.Finally,an improved random walk with restart is implemented on the heterogeneous Lnc RNA-disease network to infer potential Lnc RNA-disease associations.The major contribution of this model lies in the introduction of the concept of TCS,based on which,the velocity of convergence of TCSRWRLD can be quicken effectively,since the walker can stop its random walk while the walking probability vectors obtained by it at the nodes in TCS instead of all nodes in the whole network have reached stable state.Both comparative results and case studies have demonstrated that TCSRWRLD can achieve excellent performances in prediction of potential Lnc RNA-disease associations,which imply as well that TCSRWRLD may be a good addition to the research of bioinformatics in the future.The fifth chapter of this manuscript summarizes the above two prediction models,and on this basis,puts forward a plan for the next stage of research.