Non-coding RNA Database Construction And Cancer Genomics Analysis

Non-coding RNA(ncNRA)is a kind of RNA molecule that can't be translated into a protein.ncRNA type includes long non-coding RNA(lncRNA),microRNA(miRNA)and small nucleolar RNA(snoRNA)et al.lncRNA is defined as RNA transcripts longer than 200 nucleotides and it may involve in post-transcriptional regulation,organization of protein complexes,cell-cell signalling and allosteric regulation of proteins.RNA editing is a widespread post-transcriptional mechanism that can make a single base change on specific base in an RNA transcript.RNA editing events can result in missense codon changes and modulation of alternative splicing in mRNA,and modification of regulatory RNAs and their binding sites in noncoding RNAs.snoRNA is a type of conserved non-coding RNA with a length of 60-300 nucleotides.It is located in the nucleolus or Cajal body and participates in the process of rRNA transcriptional modification.Despite the emerging knowledge about the role of snoRNAs in cancer,the expression landscape and clinical relevance of snoRNAs in cancer have not been investigated systematically.In this study,we first systematically analyzed the A-to-I editing sites in lncRNAs across human,rhesus,mouse and fly,and observed an appreciable number of RNA editing sites which can significantly impact the secondary structures of lncRNAs and lncRNA–miRNA interactions.Then,we characterised snoRNA expression from TCGA cancer genomic data,and systematically analyzed clinical relevance of snoRNA.We obtained the following results.First,through literature review and database search,we collected lncRNA annotation information and RNA editing sites of human,mouse,rhesus and fly,predicted the impact of RNA editing on lncRNA function,and constructed database LNCediting(http://bioinfo.life.hust.edu.cn/LNCediting/).In LNCediting,there are 2,617,167 RNA editing sites and 445,599 lncRNAs.We then mapped the coordinates of the editing sites to lncRNA,and identified 191,991,1,922,165 and 1,829 editing sites in human,mouse,rhesus,and fly lncRNAs,respectively.In humans,the number of the editing sites in an lncRNA ranges from 0 to 1,248,with a mean of 16 editing sites per lncRNA.Further,we predicted the effects of editing sites on the structure and function of lncRNA and found that 123,950(human),1,585(mouse),148(rhesus)and 1,454(fly)editing sites that could affect the secondary structure of lncRNA.The number of RNA editing in lncRNA resulting in loss and gain of potential miRNA binding sites was 109,788/114,814(human),819/956(mouse),52/50(rhesus),and 157/160(fly),respectively.In addition,we provide two online tools to predict the effects of RNA editing on lncRNA secondary structure and lncRNA-miRNA interactions in LNCediting to facilitate analysis of functional effects of novel editing sites in lncRNAs.Second,in order to systematically study the expression profile and clinical relevance of snoRNA in cancer,we developed a computational pipeline and characterized the 465 snoRNA with expression level RPKM larger than 1 in 31 cancer types from TCGA miRNA-seq data.We then constructed a database named SNORic(http://bioinfo.life.hust.edu.cn/SNORic/)to store and display the snoRNA expression data of 31 cancer types.It provides snoRNA correlation analysis with tumor survival,clinical subtype,DNA methylation,copy number variation(CNV)and RNA splicing.In addition,we systematically analyzed the correlation of snoRNA expression with snoRNA host genes,CNV,and DNA methylation.We observed that snoRNAs in non-coding genes had higher expression levels than snoRNAs in protein-coding genes and snoRNAs without host genes(Kruskal-Wallis test,p = 4.83 × 10-9)and they were more enriched in non-coding genes than in protein-coding genes(p-value = 0.011).The influence of CNV and DNA methylation on regulation of snoRNA expression ranged from 0.2% to 40% in different cancer types.More importantly,we analyzed the clinical relevance of snoRNA.The tumor subtypes classified by snoRNA were found to be highly consistent with the clinical subtype of the tumor and other molecular subtype.It shows that the expression of snoRNA can be used as another dimension of cancer molecular subtyping.In addition,we screened 46 clinically relevant snoRNAs,of which SCARNA5,SNORA71 B and SNORD46 were highly negatively correlated with FOXO3.Through permutation testing,it was found that these 46 clinically relevant snoRNAs may play an important role in the development of cancer.Finally,we verified that the konckdown of SNORD46 significantly reduced the proliferation and migration of lung and breast cancer cells,demonstrating that it does play a role in cancer.In conclusion,in this study,we predict the effects of RNA editing on the lncRNA secondary structure and the interaction with miRNAs and constructed LNCEditing database which provides a reliable and unique data source for lncRNA editing.By investigating the snoRNA cancer genome and clinical relevance,we found that the expression of snoRNA was related to the clinical classification of tumors,and an snoRNA was experimentally verified to affect the proliferation and migration of cancer cells.Our study highlights the significant roles of snoRNAs in the development and implementation of biomarkers or therapeutic targets for cancer and provides a valuable resource for cancer research.