综合分析TCGA与GEO甲基化数据鉴定胆管癌预后相关基因SOX9和FZD10  

Identification of the prognostic genes SOX9 and FZD10 for cholangiocarcinoma by comprehensive analysis of DNA methylation data in TCGA and GEO

作　　者：范宁[1] 代增强 王守光 辛洋[1] 张勇[1] 孙延东[1] 臧运金[1] FAN Ning;DAI Zeng-qiang;WANG Shou-guang;XIN Yang;ZHANG Yong;SUN Yan-dong;ZANG Yun-jin(Liver Center,Affiliated Hospital of Qingdao University,Qingdao,Shandong 266000,China)

机构地区：[1]青岛大学附属医院肝病中心,山东青岛266000

出　　处：《肝胆胰外科杂志》2021年第1期25-31,共7页Journal of Hepatopancreatobiliary Surgery

摘　　要：目的鉴定胆管癌(cholangiocarcinoma,CCA)甲基化与表达谱综合生物标志物,预测CCA患者预后。方法从癌症基因组图谱(The Cancer Genome Atlas,TCGA)下载33例CCA样本和8例正常样本基因组甲基化数据及临床信息表达谱数据,同时从基因表达综合数据库(Gene Expression Omnibus,GEO)下载甲基化数据进行验证。鉴定差异甲基化基因(DMGs)与差异表达基因(DEGs)的交集基因,采用Cox比例风险回归模型鉴定甲基化生物标志物,并使用ROC曲线来评估该模型的性能。在验证组中对该模型进行验证,通过GO功能注释,探讨DNA甲基化标志物的生物学功能。结果通过对TCGA甲基化数据分析,共鉴定出600个差异甲基化基因和6876个差异表达基因,并从中筛选出与生存相关的2个甲基化基因(SOX9和FZD10),最终将SOX9和FZD10组合基因构建预后预测模型,作为CCA预后的生物标志物。ROC曲线下面积(AUC)为0.90。SOX9和FZD10组合生物标志物能够将CCA患者区分为高风险组和低风险组,低风险组患者总生存期明显高于高风险组(2.07年vs 0.92年)。多因素Cox回归分析表明,SOX9和FZD10组合生物标志物是CCA患者预后的独立预测因子。基因本体(GO)功能分析表明,SOX9和FZD10参与转录因子、转录调控、肿瘤蛋白多糖调节和干细胞的调控。结论本研究经过多组学分析,在TCGA数据中筛选出SOX9和FZD10基因组合的甲基化预后标志物,可以将CCA患者分为高风险组与低风险组,并且该组合基因是CCA独立的预后预测因子。Objective To identify the combined biomarkers of methylation and expression profile in cholangiocarcinoma(CCA)patients and to predict the prognosis of CCA.Methods DNA methylation profile data and gene expression data from 33 CCA and 8 normal tissues were downloaded from The Cancer Genome Atlas database(TCGA),and the methylation data were downloaded from Gene Expression Omnibus(GEO)as test data.The intersection genes of differential methylation genes(DMGs)and differential expression genes(DEGs)were identified.The Cox proportional hazard regression model was used to identify the methylation biomarkers,and the time-dependent ROC curve was used to evaluate the performance of the model.The model was validated,and the biological functions of DNA methylation markers were discussed by GO functional annotation.Results A total of 600 DMGs and 6876 DEGs were identified from TCGA,and two methylated genes related to survial SOX9 and FZD10 were screened out.Then SOX9 and FZD10 were used to construct a prognostic biomarker for CCA.The area under ROC curve(AUC)was 0.90.The combined biomarker of SOX9 and FZD10 could divide CCA patients into high-risk group and low-risk group.The overall survival time of low-risk group was significantly higher than that of high-risk group(2.07 years vs 0.92 years).Multivariate Cox regression analysis showed that combined biomarker SOX9 and FZD10 was independent predictor for CCA.Gene ontology(GO)functional analysis showed that SOX9 and FZD10 were involved in transcription factors,transcriptional regulation,tumor proteoglycan regulation and stem cell regulation.Conclusion In this study,by multi omics analysis,the combined biomarker SOX9 and FZD10 is identified,which can divide CCA patients into high-risk group and low-risk group.The combined biomarker SOX9 and FZD10 is independent prognostic factor for CCA.

关 键 词：胆管癌 甲基化生物标志物 癌症基因组图谱(The Cancer Genome Atlas TCGA) 基因表达综合数据库(Gene Expression Omnibus GEO) SOX9基因 FZD10基因 基因本体(GO) 

分 类 号：R735.8[医药卫生—肿瘤]