摘要
目的利用癌症基因组图谱(TCGA)数据库构建胆管癌自噬相关基因(ARGs)预后预测模型。方法通过人类自噬数据库和分子特征数据库获得531个胆管癌ARGs。从TCGA数据库中选择CHOL队列的转录组和临床数据进行下载,包含胆管癌组织36例、正常胆管组织9例。采用Perl软件将原始测序数据进行合并,提取所有ARGs的表达数据;利用R软件对胆管癌组织和正常胆管组织中的ARGs进行差异表达分析,筛选出胆管癌组织中表达失调的ARGs,并进行GO功能富集和KEGG信号通路分析。利用单因素Cox及Lasso回归模型筛选关键ARGs,多因素Cox回归模型建立预后预测模型,根据关键ARGs的mRNA表达水平和风险系数计算每个患者的风险评分,按其中位数将胆管癌患者分为高风险组和低风险组,绘制生存曲线,比较两组的生存期。绘制预后预测模型的ROC曲线,评价其预测预后的敏感度和特异性。最后利用R软件构建基于关键ARGs的列线图,绘制校准曲线评估实际生存和预测生存的一致性。结果与正常胆管组织比较,胆管癌组织中有324个表达失调的ARGs。这些ARGs主要涉及自噬、利用自噬机制的过程、大自噬、自噬的调节、凋亡等生物学过程和信号通路。经过单因素Cox和Lasso回归分析,筛选出5个关键ARGs,即VPS11、EVA1A、BNIP3、GABARAP、VPS4B。以这5个关键ARGs建立预测胆管癌患者预后的风险模型,风险评分=(-3.739×VPS11)+(1.691×EVA1A)+(1.734×BNIP3)+(5.776×GABARAP)+(-1.310×VPS4B)。生存分析显示,高风险组的总生存时间低于低风险组,预测1年、2年、3年生存率的ROC曲线下面积均大于0.9。构建了基于5个ARGs的列线图(C指数为0.822,95%CI为0.721~0.924),另外,绘制预测1年、2年、3年生存率的校准曲线几乎都落在了45°的对角线上,提示该模型的准确性和区分能力较好。结论成功构建了基于VPS11、EVA1A、BNIP3、GABARAP、VPS4B共5个关键ARGs表达的胆管癌预�
Objective To construct a survival model for predicting the prognosis of patients with cholangiocarcinoma(CCA)based on autophagy-related genes(ARGs)of The Cancer Genome Atlas(TCGA)database.Methods A total of 531 ARGs were obtained from the Human Autophagy Database and Molecular Signatures Database.The original expression profiles and corresponding clinical data of CCA patients were downloaded from the CHOL cohort of the TCGA database.There were 36 cases of cholangiocarcinoma and 9 cases of normal bile duct.Perl software was used to merge the original sequencing data and extract the expression data of all ARGs.Differential expression analysis of CCA tissue and normal tissue was performed using R software to screen out ARGs with aberrant expression.The GO functional enrichment and KEGG signaling pathway analysis were carried out using R software.ARGs were submitted to Lasso and univariate Cox regression analyses to remove the genes which might not be related to the prognosis of CCA patients.Multivariate Cox regression model was used to establish the prognostic model and we calculated the risk score of each patient according to the mRNA expression and risk coefficient of key ARGs.CCA patients were divided into the high-risk group and low-risk group according to the median of risk scores.The Kaplan Meier survival curve was plotted to analyze the median survival time.Time-dependent receiver operating characteristic(ROC)curve was drawn to investigate the sensitivity and specificity of the model.Finally,R software was used to construct the nomogram based on the key ARGs.The calibration curve was plotted to evaluate the consistency between the actual survival and the predicted survival.Results Compared with the normal tissues,there were 324 ARGs with aberrant expression in CCA tissues,and these ARGs were mainly involved in autophagy,process utilizing autophagic mechanism,macroautophagy,regulation of autophagy,apoptosis,etc.After univariate Cox and Lasso regression analysis,five key ARGs were selected,namely VPS11,EVA1A,BN
作者
史华帝
左瑜芳
钟富兰
易小琼
徐祖敏
SHI Huadi;ZUO Yufang;ZHONG Fulan;YI Xiaoqiong;XU Zumin(The Affiliated Hospital of Guangdong Medical University,Zhanjiang 524000,China)
出处
《山东医药》
CAS
2021年第2期6-11,共6页
Shandong Medical Journal
基金
广东省自然科学基金(2020A1515010048)
广东省中医药管理局科研项目(20201179)
湛江市非资助科技攻关项目(2019B01021)。
