应用生物信息学筛选胃癌诊断和预后的生物标志物

Screening Biomarkers for Diagnosis and Prognosis of Gastric Cancer by Bioinformatics

目的:应用生物信息学方法分析和筛选与胃癌诊断和预后相关的生物标志物。方法:从GEO数据库下载胃癌的基因表达谱数据集GSE79973和GSE103236。通过在线工具GEO2R和韦恩图筛选两数据集重叠的差异表达基因(DEGs)。利用仙桃在线数据平台对DEGs进行GO和KEGG富集分析。通过STRING在线工具和Cytoscape软件构建DEGs的蛋白互作网络和识别hub基因。最后,使用GEPIA、仙桃、Kaplan-Meier Plotter在线数据平台对hub基因进行表达差异分析、受试者工作特征(ROC)曲线分析及生存分析。结果:GSE79973和GSE103236两数据集中有156个重叠DEGs,包括98个上调基因和58个下调基因。其中,上调差异表达基因(uDEGs)的GO富集分析主要与细胞外基质及胶原蛋白相关;KEGG富集分析与细胞外基质受体相互作用有关。通过STRING在线工具和Cytoscape软件从重叠DEGs中识别出10个hub基因,均为uDEGs。利用GEPIA、仙桃、Kaplan-Meier Plotter在线数据平台分析表明,hub基因在胃癌组织中均显著上调(P<0.01),有一定诊断价值(AUC>0.84),并预示其预后不良(P<0.01)。结论:COL1A1、BGN、SPARC、MMP14、LOX、THBS2、TIMP1、SPP1、VCAN、COL5A2可能是胃癌诊断和预后不良的潜在生物标志物。

Objective:To analyze and screen biomarkers associated with the diagnosis and prognosis of gastric cancer by bioinformatics methods.Methods:Gene expression profiling datasets GSE79973 and GSE103236 for gastric cancer were downloaded from the GEO database.The differentially expressed genes(DEGs)overlapping the two datasets were screened by online tools GEO2R and Venn diagram.The DEGs were analyzed by GO and KEGG enrichment using the Xiantao online data platform.Using STRING online tool and Cytoscape software to construct DEGs protein interaction network and identify hub genes.Finally,GEPIA,Xiantao,and Kaplan-Meier Plotter online data platforms were used to perform differential expression analysis,receiver operating characteristic(ROC)curve analysis,and survival analysis of the hub gene.Results:There were 156 overlapping DEGs in the GSE79973 and GSE103236 datasets,including 98 up-regulated genes and 58 down-regulated genes.Among them,the GO enrichment analysis of up-regulated differentially expressed genes(uDEGs)was mainly associated with extracellular matrix and collagen;the KEGG enrichment analysis was related to extracellular matrix receptor interactions.Ten hub genes were identified from the overlapping DEGs by STRING online tool and Cytoscape software,all of which were uDEGs.Analysis using the GEPIA,Xiantao,and Kaplan-Meier Plotter online data platforms showed that the hub gene was significantly upregulated in gastric cancer tissues(P<0.01),had some diagnostic value(AUC>0.84),and predicted a poor prognosis(P<0.01).Conclusion:COL1A1,BGN,SPARC,MMP14,LOX,THBS2,TIMP1,SPP1,VCAN,COL5A2 may be potential biomarkers for the diagnosis and poor prognosis of gastric cancer.