Analysis Of MiRNA-Target Gene Pairs In Pan-Cancers Based On TCGA Database

Cancer is one of the major diseases that threaten human health throughout the world.It involves complex molecular regulation mechanisms in tumorigenesis,development and migration.MicroRNAs(miRNAs)are defined as small regulatory RNA molecules with 19-24 nucleotides in length.They repress gene expression by binding to complementary sequences in the 3' untranslated region(3' UTR)of mRNAs to target them for degradation and thereby prevent their translation.Accumulating evidences demonstrated that microRNA-target gene pairs were closely related to tumorigenesis and development.Deepening the study of miRNA-target genes in cancer will help us to reveal the molecular mechanisms of tumorgenesis more clearly.In this study,we integrated miRNA and mRNA expression profiles of 11 common cancer types in The Cancer Genome Atlas(TCGA)database.The 11 cancer types were bladder urothelial carcinoma,breast invasive carcinoma,head and neck squamous cell carcinoma,kidney chromophobe,kidney renal clear cell carcinoma,kidney renal papillary cell carcinoma,liver hepatocellular carcinoma,lung adenocarcinoma,lung squamous cell carcinoma,stomach adenocarcinoma,and thyroid carcinoma.We firstly screened the differentially expressed miRNAs and genes in tumor compared with in normal.Then,we obtained the miRNA-target gene pairs and analyzed their co-expression coefficients in normal tissues and tumor tissues respectively.Finally,we performed GO and KEGG functional annotations on these miRNA-target gene pairs with regulatory differences between normal and tumor tissues.Additionally,we screened the several-genes signature for cancer prognosis prediction and performed survival analysis for high risk group and low risk group cancer patients.The details are as follows:(1)We collected the miRNA and mRNA expression profiles of 11 cancer types from TCGA database,and obtained the differentially expressed miRNAs and genes in tumor by using DESeq2.The thresholds were set as |log2FC| > 1 and padj(adjust p value)< 0.05 to select differentially expressed miRNAs or genes.We totally identified 62 ? 176 differentially expressed miRNAs and 1 483 ? 4 875 differentially expressed genes for the 11 cancer types.By summaring differentially expressed miRNAs and genes from 11 cancer types,we totally identified 496 differentially expressed miRNAs and 12 037 differentially expressed genes,and 51.6% of miRNAs and 77.8% of genes were differentially expressed in multicancers.Additionally,we identified 143 miRNAs and 4 737 genes,which showed discordantly regulated status in different cancer types.For example,MYBL2 was downregulated in thyroid carcinoma but upregulated in the remaining 10 cancer types.The results showed that there was a wide range of differentially expressed miRNAs and genes in each cancer type.And there are huge heterogeneities between different cabcer type and also some common characters.(2)To obtain miRNA-target gene pairs for each cancer type,we integrated differentially expressed miRNAs and genes with experimentally validated miRNA-target gene relationships collected by two databases,miRTarBase and miRecords.We calculated the Pearson correlation coefficient of miRNA-target gene pairs in normal tissue samples and tumor tissue samples seperatively for each cancer type.The results showed that these miRNA-target gene pairs with strongly correlation(Pearson correlation coefficient <-0.6,P < 0.05)in normal tissues and these miRNA-target gene pairs with strongly correlation in tumor were greatly different.We also found that miRNA-target gene pairs showed a significantly lower correlation in tumor tissue than that of in normal.(3)To gain global insights into function of miRNA-target gene pairs that showed different regulate status between normal and tumor,we performed GO and KEGG annotation by using DAVID website database.These dysregulated miRNA-target were found to be mostly participant in biological processes and signaling pathways,such as focal adhesion,p53,MAPK,and PI3K-Akt signaling pathways,who were hubs pathways in cancer progression,invasion,and metastasis.In addition,to pick the prognostic related genes,robust likelihood-based survival analysis was performed by using Rwith Rbsurv package.The risk score formula was established by multivariate Cox regression analysis,and cancer patients were divided into high-risk group and low-risk group based their risk score.K-M survival analysis showed that there was a significant difference in survival time between high-risk and low-risk cancer patients.Conclusion: Differential analysis showed that there was a wide range of differentially expressed miRNAs and genes for each cancer type.There were huge heterogeneities between different cancer types,but also some commonalities.The strongly regulated miRNA-target genes pairs were mostly different between normal and tumor.The correlation coefficient of miRNA-target gene pairs among tumor samples was generally smaller than that of them among normal samples.Functional analysis showed that the differentially regulated miRNA-target gene pairs between normal and tumor play an important role in tumorigenesis,metastasis,and prognosis.