基于TCGA数据库应用生物信息学方法分析和挖掘肺腺癌预后和诊断miRNA研究

目的基于TCGA数据库,应用生物信息学方法分析和挖掘肺腺癌预后和诊断miRNA生物学标志物。方法数据下载:从TCGA下载获取肺腺癌miRNA表达谱数据,包括miRNAseq和临床数据。筛选差异表达miRNAs:应用R-version 3.6.2软件中的edgeR包筛选肺腺癌组织和正常肺组织的差异基因,以│logFC│>2,P<0.05为筛选条件。筛选与预后相关miRNAs:应用R-version 3.6.2软件中的survival包绘制KM生存曲线,筛选与预后相关的miRNAs。筛选可作为肺腺癌诊断的miRNAs:应用R-version 3.6.2软件中的pROC包制作受试者工作特征曲线(ROC)评价与预后相关miRNAs诊断肺腺癌的特异性和敏感性。结果共识别到肺腺癌与正常肺组织差异表达的miRNA 144个,其中上调表达119个,下调表达25个。通过K-M生存曲线筛选出与预后显著相关(P<0.05)的miRNA共13个,分别为hsa-miR-139-3p、hsa-miR-328-3p、hsa-let-7g-3p、hsa-miR-142-3p、hsa-miR-147b、hsa-miR-31-5p、hsa-miR-548v、hsa-miR-188-3p、hsa-miR-31-3p、hsa-miR-490-3p、hsa-miR-4664-3p、hsa-miR-1293、hsa-miR-615-3p,其中hsa-miR-147b、hsa-miR-4664-3p、hsa-miR-1293、hsa-miR-615-3p与预后呈负相关,其他呈正相关。应用ROC评价上述与预后相关miRNAs诊断肺腺癌的特异性和敏感性,其中hsa-miR-147b、hsa-miR-142-3p、hsa-let-7g-3p、hsa-miR-139-3p 4个miRNA的AUC曲线>0.9,分别为0.904、0.927、0.945、0.965,提示具有较高诊断价值及准确性。结论 hsa-miR-147b、hsa-miR-142-3p、hsalet-7g-3p、hsa-miR-139-3p可作为肺腺癌预后评估和诊断miRNA的生物学标志物,具有重要的临床价值。

Analysis and mining of bioinformatics based on TCGA database miRNA study on prognosis and diagnosis of lung adenocarcinoma

Objective To analyze and mine miRNA biomarkers for prognosis and diagnosis of lung adenocarcinoma based on TCGA database.Methods Data download, miRNA expression profile data of lung adenocarcinoma, including miRNAseq and clinical data, were obtained from TCGA download. Screening differentially expressed miRNAs, differential genes in lung adenocarcinoma and normal lung tissues were screened by edgeR package in R-version 3.6.2 software. The screening criteria was│logFC│>2, P<0.05. Screening miRNAs related to prognosis, KM survival curve was made by survival package in R-version 3.6.2 software, and miRNAs related to prognosis was screened. Screen the miRNAs that could be used for the diagnosis of lung adenocarcinoma, the receiver operating characteristic curve(ROC) was made by using pROC package in R-version3.6.2 software to evaluate the specificity and sensitivity of miRNAs in the diagnosis of lung adenocarcinoma. Results 144 miRNA differentially expressed between lung adenocarcinoma and normal lung tissues were identified. 119 miRNAs were up-regulated and 25 miRNAs were down-regulated.A total of 13 miRNA were screened out from K-M survival curve, which were significantly related to prognosis, they were hsa-miR-139-3 p, hsa-miR-328-3 p, hsa-let-7 g-3 p, hsa-miR-142-3 p, hsa-miR-147 b, hsa-miR-31-5 p, hsa-miR-548 v, hsa-miR-188-3 p, hsa-miR-31-3 p, hsa-miR-490-3 p, hsa-miR-4664-3 p, hsa-miR-1293 and hsa-miR-615-3 p. Among them, hsa-miR-147 b, hsa-miR-4664-3 p, hsa-miR-1293, hsa-miR-615-3 p were negatively correlated with prognosis, while others were positively correlated. The specificity and sensitivity of miRNAs related to prognosis in the diagnosis of lung adenocarcinoma were evaluated by ROC. The AUC of hsa-miR-147 b, hsa-miR-142-3 p, hsa-let-7 g-3 p and hsa-miR-139-3 p miRNA were more than 0.90,which were 0.904, 0.927, 0.945 and 0.965, respectively. It was suggesting that they have high diagnostic value and accuracy. Conclusion hsa-miR-147 b, hsa-miR-142-3 p, hsa-let-7 g-3 p and hsa-miR-139-3 p can be used as miRNA biomarkers for prognosis evaluation and diagnosis of lung adenocarcinoma and have important clinical value.