基于高血压相关基因构建腹主动脉瘤破裂的分子诊断模型  

Construction of molecular diagnostic model for ruptured abdominal aortic aneurysm based on hypertension related genes

作　　者：化召辉[1] 周保宁 张帅[1] 张林枫[1] 岳永强[1] 李震[1] Hua Zhaohui;Zhou Baoning;Zhang Shuai;Zhang Linfeng;Yue Yongqiang;Li Zhen(Department of Endovascular Surgery,the First Affiliated Hospital of Zhengzhou University,Zhengzhou 450052,China)

机构地区：[1]郑州大学第一附属医院腔内血管外科,郑州450052

出　　处：《中华实验外科杂志》2023年第8期1590-1592,共3页Chinese Journal of Experimental Surgery

基　　金：2023年度河南省重点研发与推广专项(科技攻关)项目(232102311033)。

摘　　要：目的通过生物信息学分析方法来识别腹主动脉瘤破裂相关基因与高血压差异基因, 筛选出共同关键基因并建立腹主动脉瘤破裂的分子诊断模型。方法所有样本数据均从基因表达综合数据(GEO)下载。对36例样本(高血压组28例, 对照组8例, 2015年由法国图卢兹大学心脏病研究所测序获得)的全转录组数据进行差异分析以得到差异基因;对48例腹主动脉瘤样本(17例破裂, 31例未破裂, 2017年由德国慕尼黑大学检验医学研究所测序获得)的转录组数据进行加权基因共表达网络分析(WGCNA)来识别与腹主动脉瘤破裂最为相关的基因模块。随后对得到的基因取交集并利用机器学习算法(Lasso回归)筛选出腹主动脉瘤破裂的关键基因。利用关键基因构建诊断模型并采用受试者工作特征(ROC)曲线计算诊断模型的ROC曲线下面积(AUC)。该研究中利用未配对的学生t检验和Wilcoxon检验来比较两组间的差异。结果对高血压组及对照组样本进行差异分析共得到444个差异基因。对腹主动脉瘤样本利用WGCNA显示当软阈值为7时符合无尺度网络, 淡绿色模块正相关性最高, 包含848个基因。将444个基因与848个基因取交集共得到28个基因。利用Lasso回归对28个基因进行筛选, 共得到4个关键基因(MAP2K1、KLF9、UTP4、RHOU)。利用这4个基因构建诊断模型来预测患者是否容易发生动脉瘤破裂。利用ROC曲线评估该诊断模型预测腹主动脉瘤破裂的效能, 发现曲线下面积为0.825。结论利用MAP2K1, KLF9, UTP4, RHOU构建的诊断模型有助于腹主动脉瘤破裂的预测。Objective To identify genes associated with ruptured abdominal aortic aneurysm and hypertension differential genes by bioinformatics analysis,to screen out common key genes and to establish a molecular diagnostic model for ruptured abdominal aortic aneurysm.Methods All sample data were downloaded from the gene expression omnibus(GEO).Differential analysis was performed on whole transcriptome data from 36 samples(28 in the hypertension group and 8 in the control group,sequenced by the Institute of Cardiology,University of Toulouse,France,in 2015)to obtain differential genes;weighted gene co-expression network analysis(WGCNA)was performed on RNA expression profile data from 48 abdominal aortic aneurysm samples(17 ruptured and 31 unruptured,sequenced by the Institute of Laboratory Medicine,University of Munich,Germany,in 2017)to identify the gene modules most associated with ruptured abdominal aortic aneurysm.The obtained genes were then intersected and screened for key genes for ruptured abdominal aortic aneurysm using a machine learning algorithm(Lasso regression).Diagnostic models were constructed using the key genes and subject operating characteristic(ROC)curves were used to calculate the AUC values of the diagnostic models.The t-test and Wilcoxon test of unmatched students in the study were used to compare the differences between the two groups.Results A total of 444 differential genes were obtained by differential analysis of samples from the hypertensive and control groups.For the abdominal aortic aneurysm samples WGCNA showed compliance with the scale-free network when the soft threshold was 16,and the light green module had the highest positive correlation and contained 848 genes.A total of 28 genes were obtained by taking the intersection of 444 genes with 848 genes.The 28 genes were screened using Lasso regression,and a total of 4 key genes(MAP2K1,KLF9,UTP4,RHOU)were obtained,which were used to form a diagnostic model to predict whether patients were prone to aneurysm rupture.The efficacy of this diagnos

关 键 词：腹主动脉瘤 高血压 关键基因 机器学习 

分 类 号：R543.16[医药卫生—心血管疾病] R544.1[医药卫生—内科学]