Research On Feature-based LncRNA Set Comparison And Similarity Calculation

Long non-coding RNAs(lncRNAs)is a class of long(>200 nucleotides)transcripts produced from non-protein coding gene.Although lncRNAs do not directly achieve biological functions by encoding proteins,many studies have shown that lncRNAs are widely found in various stages of biological processes and many diseases,playing crucial regulatory roles.Nevertheless,the fraction of lncRNAs with validated functions is still limited,mainly due to the time-and cost-intensive nature of experimental assay.On the other hand,accumulating reports have proposed various computational models to extensively predict lncRNAs function based on its interaction with other genes and/or its known associations with diseases and the sophisticated algorithms.One principle of this methods is that similar genes should have similar function,so the study on the similarity of Lnc RNAs was also carried out in depth.However,how to collect and integrate various biological features,the fundamental issue for similarity measurement,was less discussed.To this end,here we integrated multiple type lncRNAs features to establish a feature-based lncRNAs set comparison and lncRNAs similarity calculation method.While start from the fundamental biological features of lncRNAs,this study would provide a helpful indication for the methodology of lncRNAs function studies.More specifically,in this study,by referring to the literature on lncRNAs researches in recent years,we first summarized and analyzed the methods proposed in several studies.Combined with the methods and strategies for studying the functions of coding genes and other non-coding RNAs,a computational method scheme was proposed to study lncRNAs and their similarity from the perspective of gene feature information including genomic features,cellular expression features,tissue expression features,subcellular localization features and phenotypic association features.By integrating multiple-type lncRNAs features,the proposed method could not only compare lncRNAs gene set according to their associated features,but also calculate lncRNAs similarity based on the biological features.This study is expected to explore a new research direction for the researches of lncRNAs by starting from the most basic feature information of genes in the absence of effective methods to study lncRNAs related issues.To facilitate researchers to use the method designed in this paper,we designed and developed a website system to build an online platform LnCompare(http://www.rnanut.net/lncompare).The platform implemented the above functions of feature-based lncRNAs set comparison and lncRNAs similarity calculation.At the same time,with the help of the data visualization scheme,the analysis results can be visualized in various user-friendly fashions.In addition,in order to further facilitate researches on the basis of the platform,the platform provides data download function,where users can download the data freely according to their needs.