基于WGCNA和LASSO分析的膀胱癌生物学标志物鉴定  

Identification of Biomarkers for Bladder Cancer Based on WGCNA and LASSO Analyses

作　　者：王书恒 刘薇薇 李九智[1] WANG Shuheng;LIU Weiwei;LI Jiuzhi(Department of urology,Xinjiang Uygur Autonomous Region People′s Hospital,Xinjiang 830002,China)

机构地区：[1]新疆维吾尔自治区人民医院泌尿科,乌鲁木齐830002 [2]乌鲁木齐市口腔医院种植科,830002

出　　处：《医学研究杂志》2025年第3期54-61,89,共9页Journal of Medical Research

基　　金：新疆维吾尔自治区科技厅自然科学基金青年基金资助项目(2022D01C823)。

摘　　要：目的膀胱癌(bladder cancer,BLCA)是常见的疾病,其发病机制不清楚,本研究旨在发现膀胱癌的关键基因,为今后的预防和治疗提供依据。方法从美国生物技术信息中心NCBI的GEO(Gene Expression Omnibus)数据库中获取膀胱癌数据集GSE121711,对GEO数据进行加权基因共表达网络分析(weighted gene coexpression network analysis,WGCNA)鉴定样品中与BLCA高度相关的基因模块,提取差异表达基因(differentially expressed gene,DEG)和模块中基因的交集基因,交集基因做基因本体论(gene ontology,GO)和京都基因和基因组百科全书(Kyoto Encyclopedia of Genes and Genomes,KEGG)分析,进一步通过蛋白互作(protein-protein interaction,PPI)网络筛选degree值最高的关键基因,之后进行聚类分析。最后,利用LASSO建立诊断模型。采用反转录实时定量聚合酶链式反应(RT-qPCR)检测枢纽基因在BLCA组织与正常组织中的表达情况。结果WGCNA显示黑色模块与膀胱癌的关联性最显著,黑色模块中有611个基因并与DEG取交集,得到449个共有基因。利用LASSO构建由RAC3、APOL4、FASN和CLASRP组成的诊断模型,利用受试者工作特征(receiver operating characteristic,ROC)曲线进行了365天(1年)、1095天(3年)、1825天(5年)时间点的分析,曲线下面积(area under curve,AUC)分别为80%、82%、85%。并在GSE101723和GSE83586的合并数据集上进行验证,发现结果与生物信息学的结果相似。BLCA组织中枢纽基因RAC3、APOL4、FASN和CLASRP mRNA相对表达量显著高于正常组织(t=8.074,P<0.0001;t=3.577,P<0.001;t=12.241,P<0.0001;t=8.846,P<0.0001)。结论本研究构建了BLCA诊断模型,发现RAC3、APOL4、FASN和CLASRP是潜在的生物学标志物,可能为提高膀胱癌的早期诊断和治疗提供新的依据。Objective Bladder cancer(BLCA)is a common disease,and the pathogenesis of which is not clear.This study aims to find the key genes of bladder cancer for future prevention and treatment.Methods The bladder cancer dataset GSE121711 was obtained from the Gene Expression Omnibus(GEO)database of NCBI.The weighted gene coexpression network analysis(WGCNA)was performed on GEO data to identify the gene modules highly associated with BLCA in the samples.The intersecting genes of differentially expressed genes(DEG)and genes in the module were extracted.The common genes were analyzed by Gene Ontology(GO)and Kyoto Encyclopedia of Genes and Genomes(KEGG),and the key genes with the highest degree were further screened through the Protein-protein interaction(PPI)network.The cluster analysis is carried out.Finally,the LASSO is used to establish the diagnostic model.The expression of hub genes in BLCA tissues and normal tissues was detected by using reverse transcription real-time quantitative polymerase chain reaction(RT-qPCR).Results WGCNA showed the most significant association between the black module and bladder cancer.There were 611genes in the black module and intersected with DEG for 449 common genes.A diagnostic model consisting of RAC3,APOL4,FASN and CLASRP was constructed using LASSO,and analysis was conducted using receiver operating characteristic(ROC)curves at time points of 365 days(1 year),1095 days(3 years)and 1825 days(5 years).The Area Under Curve(AUC)of 365(1 year),1095(3 years)and 1825(5 years)were 80%,82%and 85%,respectively.The results were verified on the combined dataset of GSE101723 and GSE83586,which were found to be similar to those of bioinformatics.The relative expression levels of hub genes RAC3,APOL4,FASN and CLASRP mRNA in BLCA tissues were significantly higher than those in normal tissues(t=8.074,P<0.0001;t=3.577,P<0.001;t=12.241,P<0.0001;t=8.846,P<0.0001).Conclusion We constructed a BLCA diagnostic model and found that RAC3,APOL4,FASN and CLASRP were potential biomarkers that may provide new ins

关 键 词：膀胱癌 生物信息学 加权基因共表达网络分析 生物学标志物 

分 类 号：R737.14[医药卫生—肿瘤]