论文部分内容阅读
目的:应用生物信息学方法筛选胰腺导管腺癌的预后风险标志物。方法:从TCGA数据库下载胰腺导管腺癌患者的临床资料、miRNA和基因表达谱数据。然后应用弹性网络Cox比例风险回归(EN-Cox)模型、受试者工作特征(ROC)曲线和生存分析筛选出与胰腺导管腺癌预后风险明显相关的miRNA和基因。最后,对筛选到的预后风险基因与miRNA的潜在靶基因进行文献挖掘及功能分析。结果:经过数据预处理,共得到137例胰腺导管腺癌患者的完整临床资料及797个miRNA和19 969个基因表达谱数据。基于λ=0.107的参数值,EN-Cox分析筛选出了包括54个基因和5个miRNA在内的59个潜在的预后风险因素;根据ROC曲线确定病例分组的截断值,并绘制Kaplan-Meier曲线,最后共筛选出17个胰腺导管腺癌预后风险标志物(均P<0.05),包括16个基因和1个miRNA(miRNA-125a)。在16个预后风险基因中,谷胱甘肽S转移酶μ4(GSTM4)、可诱导T细胞共刺激分子配体(ICOSLG)、精子发生相关2(SPATA2)同时又是miRNA-125a的靶基因;只有GATA结合蛋白1(GATA1)为转录因子编码基因。结论:所筛选的因子在胰腺癌中的作用还有待阐明,并有望成为判断胰腺导管腺癌预后的指标及治疗靶点。
Objective: To screen the prognostic risk markers of pancreatic ductal adenocarcinoma using bioinformatics methods. Methods: Clinical data, miRNA and gene expression profiles of patients with pancreatic ductal adenocarcinoma were downloaded from the TCGA database. Then using the elastic Cox proportional hazards regression model (EN-Cox), receiver operating characteristic (ROC) curves and survival analysis, the miRNAs and genes that were significantly associated with the prognosis of pancreatic ductal adenocarcinoma were screened out. Finally, the potential prognostic genes screened and potential targets of miRNAs were searched and analyzed by literature. Results: After data preprocessing, a total of 137 cases of pancreatic ductal adenocarcinoma patients with complete clinical data and 797 miRNA and 19 969 gene expression profiling data. Based on the parameter values of λ = 0.107, 59 potential risk factors including 54 genes and 5 miRNAs were screened out by EN-Cox analysis. According to the ROC curve, the cut-off value of the case group was determined and the Kaplan-Meier curve Finally, 17 prognostic markers of pancreatic ductal adenocarcinoma were screened (all P <0.05), including 16 genes and 1 miRNA (miRNA-125a). Among the 16 prognostic risk genes, glutathione S-transferase μ4 (GSTM4), inducible T cell costimulatory ligand (ICOSLG), and spermatogenesis-related gene 2 (SPATA2) are also target genes of miRNA-125a. Only GATA binding protein 1 (GATA1) is a transcription factor-encoding gene. Conclusion: The role of the selected factors in pancreatic cancer remains to be elucidated, and is expected to be an index and therapeutic target for the prognosis of pancreatic ductal adenocarcinoma.