LUSC Differentially Expressed Genes Were Analyzed Based On TCGA Database And CeRNA Regulatory Network Was Constructed

Objective:The differentially expressed LncRNAs,mRNAs and miRNAs in LUSCs were screened from normal tissue and cancer samples in the TCGA database.The ceRNA network of LUSC was established through the differentially expressed miRNAs obtained above.Bioinformatics analysis of differentially expressed genes was conducted to explore their biological functions and signaling pathways.This study contributes to the development of the molecular mechanism and tumor occurrence,lays a foundation for the prevention and diagnosis of LUSC,and also provides more basis for the discovery of new related drug therapeutic targets and molecular markers in the clinical treatment of LUSC.Method:1.Screening of differentially expressed gene data sets: the matrices of LUSC and non-Lusc tissues were downloaded from the We downloaded the gene of LUSC and non-LUSC tissues from the TCGA database,then we use perl script to merge the matrix,and differentially expressed LncRNAs,mRNAs and miRNAs were obtained using the "edge" package in R software.2.Functional enrichment analysis of mRNAs: DAVID database was used for GO and KEGG pathway analysis to achieve functional enrichment analysis of differentially expressed mRNA.The PPI was build by using STRING database,and Cytoscape3.8.0 was used for visualization and Hub gene screening by degree algorithm.3.Screening variables: We used univariate Cox regression analysis and LASSO regression analysis to search variables related to prognosis,and then risk scoring model was established by multivariate Cox regression analysis.4.Follow-up survival analysis was performed on the obtained RNAs to determine whether they could be used as independent prognostic factors.5.Use the miRNAs obtained above to establish ceRNA network.Result:1.edge R package was used to analyze LUSC and non-LUSc data of TCGA,and a total of 1940 LncRNAs were differentially expressed,among which 1348 genes were up-regulated and 592 genes were down-regulated.A total of 2775 mRNAs were differentially expressed,among which 1522 genes were up-regulated and 1253 genes were down-regulated.A total of 161 miRNAs were differentially expressed,among which 133 genes were up-regulated and 28 genes were down-regulated.2.The protein interaction network was constructed,and 10 Hub genes were screened out: FOS,AURKB,CDCA8,BUB1,CCNB1,TOP2 A,TTK,UBE2 C,ITGA1,RRM2.3.Univariate and multivariate Cox regression analysis showed that 29 LncRNAs constructed Cox models were used to predict the prognosis of patients with lung squamous cell carcinoma.7 differentially expressed miRNAs were obtained by univariate Cox regression analysis and LASSO regression: hsa-mir-139,hsa-mir-326,hsa-mir-4664,hsa-mir-4745,hsa-mir-616,hsa-mir-7641-2,hsa-mir-944.4.According to the miRNAs obtained above as mediators,a ceRNA network composed of 6 miRNAs,28 LncRNAs and 316 mRNAs was obtained.Conclusion: 1.Differentially expressed mRNAs of LUSC were screened and the Top10 Hub gene was screened to provide a potential new target for early prevention and control of LUSC.2.29 risk models constructed by LncRNAs can be used to predict the prognosis of LUSC patients.3.The risk scoring model constructed by hsa-mir-139,hsamir-326,hsa-mir-616,h SA-Mir-7642-2,hsa-mir-4664,hsa-mir-944,and hsa-mir-4745 has good accuracy.C index is 0.633(se=0.021).4.The ceRNA network,consisting of 6miRNAs,28 LncRNAs and 316 mRNAs,affects the occurrence and development of LUSCs through intermolecular regulation.