加权基因共表达网络分析结合机器学习筛选糖尿病心肌病中与Th17细胞相关的核心基因  

Identification of core genes related to Th17 cells in diabetic cardiomyopathy through weighted gene co-expression analysis combined with machine learning

作　　者：秦冲 王振辉 王越 董健 赵娜 刘擎[2] 张洪亮[2] QIN Chong;WANG Zhenhui;WANG Yue;DONG Jian;ZHAO Na;LIU Qing;ZHANG Hongliang(College of Clinical Medical,Jiamusi University,Jiamusi 154000,China;Department of Cardiology,the First Affiliated Hospital of Jiamusi University,Jiamusi 154000,China)

机构地区：[1]佳木斯大学临床医学院,黑龙江佳木斯154000 [2]佳木斯大学附属第一医院心内科,黑龙江佳木斯154000

出　　处：《现代医学》2025年第4期546-558,共13页Modern Medical Journal

摘　　要：目的:探讨免疫细胞相关基因在糖尿病心肌病(DCM)中的作用,并探讨其潜在机制。方法:采用单样本基因集富集分析探讨样本的免疫浸润情况。通过加权基因共表达网络分析(WGCNA)挖掘与免疫浸润相关的基因。通过富集分析,分析这些基因的功能。基因间的相互作用在蛋白-蛋白相互作用网络中可视化,通过Cytoscape软件鉴定核心基因。通过机器学习筛选核心基因,生成受试者工作特征(ROC)曲线来评估核心基因在DCM诊断中的表现。采用实时荧光定量逆转录聚合酶链反应(qRT-PCR)和Western blot检测核心基因在大鼠心肌组织中的表达。结果:DCM组样本中辅助性T细胞17(Th17)细胞浸润分数显著增加。差异表达分析和WGCNA鉴定出28个与Th17细胞浸润相关的差异基因,这些基因与代谢相关。使用LASSO回归、随机森林模型等算法从28个差异基因中鉴定出核心基因。此外,ROC曲线分析显示,核心基因具有极佳的区分DCM的能力。最后,体内实验表明,qRT-PCR和Western blot的结果与测序结果一致。结论:WGCNA联合机器学习筛选出DCM中与Th17细胞相关的核心基因为2,4-二烯酰辅酶a还原酶1(Decr1)和线粒体3-羟基-3-甲基戊二酰辅酶A合成酶2(Hmgcs2),在DCM中表达量均显著高于正常。Objective:To investigate the role of immune cell-related genes in diabetic cardiomyopathy(DCM)and their mechanistic underpinnings.Methods:Immune infiltration profiles were analyzed via single-sample gene set enrichment analysis(ssGSEA).Weighted gene co-expression network analysis(WGCNA)identified immune-associated gene modules.Functional enrichment analyses(GO/KEGG)elucidated biological pathways,while protein-protein interaction(PPI)networks(STRING/Cytoscape)revealed core genes.Machine learning(LASSO/random forest)screened core diagnostic biomarkers,validated by receiver operating characteristic(ROC)curves.Expression levels of core genes were experimentally confirmed in rat DCM myocardial tissues using qRT-PCR and Western blot.Results:The DCM group exhibited significantly elevated T helper cell 17(Th17)cell infiltration scores.Differential expression analysis combined with WGCNA identified 28 Th17-associated genes enriched in metabolic pathways.Machine learning algorithms(LASSO/random forest)screened core genes from this cohort,validated by ROC curves demonstrating robust DCM diagnostic capacity.Experimental validation via qRT-PCR and Western blot confirmed consistent expression patterns between sequencing data and myocardial tissues in vivo.Conclusion:Integrated WGCNA and machine learning revealed two Th17-associated core genes in DCM:Decr1(2,4-dienoyl-CoA reductase 1)and Hmgcs2(mitochondrial 3-hydroxy-3-methylglutaryl-CoA synthase 2),both showing significantly upregulated expression compared to controls.

关 键 词：糖尿病心肌病 辅助性T细胞17 加权基因共表达网络分析 机器学习 核心基因 

分 类 号：R542.2[医药卫生—心血管疾病] R587.2[医药卫生—内科学]