Construction Of CeRNA Regulatory Network And Prognositic Model For Gastric Cancer Based On TCGA Database

Malignant tumor is a life-threatening disease that causes a global economic and social burden.The incidence and mortality of gastric cancer are ranking high among malignant tumors.The prevention,pathogenesis,and prognosis of gastric cancer still demand further research.Non-coding RNA plays an important role in the occurrence and development of malignant tumors,especially in the mechanism of the interaction between long non-coding RNA and microRNAs affecting tumor development.Objective: To study the interaction between long non-coding RNAs and microRNAs and their competitive binding,which regulate the downstream mRNAs in gastric cancer transcriptome,and to construct a model which contains non-coding RNAs to guide the prognosis of gastric cancer.Methods:Download the expression data file of gastric cancer RNA group from TCGA official website,extract the expression matrix of long-chain non-coding RNA and mRNA from the RNA expression file,and analyze the differential expression of genes using R software package..The lncRNAs file was compared with the mircode database comparison file to extract the relevant lncRNA-miRNAs.Using three miRNA downstream target prediction websites,the miRDB,miRTarBase,and TargetScan,the targets for differentially expressed miRNAs were predicted.lncRNA-miRNAs and miRNA-mRNAs files were collected to build a competitive endogenous RNA(ceRNA)regulatory network.Survival curve of the key genes in the ceRNA network was drawn.The R software package was used to further analyze differentially expressed genes to construct a prognostic model for gastric cancer.Results:After collating the differentially expressed lncRNAs,miRNAs and mRNA,and further comparing with the corresponding database,the differentially expressed lncRNAs file was compared with the mircode database comparison file,and 152 pairs of related lncRNA-miRNAs were extracted and obtained in total 9 target genes.After collating the obtained data,a gastric cancer ceRNA regulatory network was constructed.Based on the results of gene survival analysis in the ceRNA regulatory network,this study constructed a prediction model of lncRNAs related to the prognosis of gastric cancer containing 4 lncRNAs.Using model evaluation methods,this study found that the prediction model had moderate predictive power.Conclusions: The study obtained the latest ceRNA regulatory network of gastric cancer,which can be further verified.This regulatory network can provide reference for the study of the mechanism of gastric cancer.The gastric cancer prognosis model has certain predictive power,which can be further verified in practice.