| 缺血性心肌病中乳酸化相关基因和免疫浸润的生物信息学分析 |
| 投稿时间：2025-05-09 修订日期：2025-07-14 点此下载全文 |
| 引用本文：龚勇,热甫开提·阿不都哈力克,段东琴,木亚沙尔·阿布都西热提,艾力曼·马合木提.缺血性心肌病中乳酸化相关基因和免疫浸润的生物信息学分析[J].医学研究杂志,2025,54(11):100-109 |
| DOI：
10.11969/j.issn.1673-548X.2025.11.019 |
| 摘要点击次数: 29 |
| 全文下载次数: 17 |
|
| 基金项目:新疆维吾尔自治区重点研发任务专项(子课题)(2022B03023-4) |
|
| 中文摘要:目的 利用生物信息学筛选缺血性心肌病(ischemic cardiomyopathy, ICM)中乳酸化相关基因(lactylation-related gene, LRG),并分析其免疫浸润模式。方法 从基因表达数据集(Gene Expression Omnibus,GEO)数据库中下载ICM相关基因芯片数据集GSE57338、GSE26887、GSE79962和GSE42955,从MSigDB数据库获取LRG,分析GSE57338数据集中的差异表达基因(differential expressed gene, DEG),对DEG进行基因本体论(Gene Ontology,GO)功能富集和京都基因与基因组百科全书(Kyoto Encyclopedia of Genes and Genomes, KEGG)信号通路富集分析。通过加权基因共表达网络分析(weighted gene coexpression network analysis, WGCNA)获取ICM相关基因模块,筛选WGCNA与LRG交集基因,确定核心基因,基于核心基因通过最小绝对收缩和选择算子(least absolute shrinkage and selection operator, LASSO)回归和随机森林(random forests, RF)算法共同筛选得到特征基因,将特征基因进行Logistic回归分析并构建列线图,绘制受试者工作特征(receiver operating characteristic, ROC)曲线,评估其对ICM的诊断价值。利用验证集对特征基因进行验证,探索免疫细胞浸润模式,构建转录因子-特征基因调控网络,并与特征基因进行相关性分析。结果 筛选GSE57338数据集得到了5709个DEG,其中表达上调的基因2817个,表达下调的基因2892个。通过两种机器学习算法筛选出与ICM相关的4个LRG特征基因,即JMJD1C、EIF4G1、PPIL4和PHC3,其构建的诊断模型能够较准确地区分ICM,决策曲线和校准曲线显示该模型具有良好的性能,外部验证集表明该模型具有较高的诊断准确率,曲线下面积(area under the curve, AUC)为0.909。特征基因富集分析(gene set enrichment analysis, GSEA)结果显示,免疫炎性反应和代谢紊乱可能在ICM的发病机制中起重要作用。结论 本研究成功筛选了4个LRG作为ICM诊断潜在生物学标志物,并建立了新的ICM诊断模型,可以较准确地区分ICM患者,为未来精准诊断与治疗提供了新的思路。 |
| 中文关键词:缺血性心肌病 乳酸化相关基因 机器学习 生物信息学 诊断模型 |
| |
| Bioinformatics Analysis of Lactatation-related Gene and Immune Infiltration in Ischemic Cardiomyopathy. |
|
|
| Abstract:Objective To screen for lactylation-related gene (LRG) in ischemic cardiomyopathy (ICM) using bioinformatics, and analyze their immune infiltration patterns. Methods ICM-related gene expression datasets GSE57338, GSE26887, GSE79962 and GSE42955 were downloaded from the Gene Expression Omnibus (GEO) database, and LRG were obtained from the MSigDB database. Differential expressed gene (DEG) in GSE57338dataset were analyzed. Gene Ontology (GO) functional enrichment and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analyses were performed on DEG. Weighted gene co-expression network analysis (WGCNA) was constructed to identify gene modules correlated with ICM. Intersecting genes between WGCNA and LRG were screened to determine core genes. Based on the core genes, the feature genes were selected via least absolute shrinkage and selection operator (LASSO) regression and random forest (RF) algorithms, followed by Logistic regression analysis to build a Nomogram. Receiver operating characteristic (ROC) curves were plotted to evaluate the diagnostic value of feature genes for ICM. Validation set was used to verify the feature genes, explore the infiltration pattern of immune cells, and transcription factor-feature gene regulatory network was constructed, and correlation between feature genes and immune cells were analyzed. Results A total of 5709 DEG were identified in GSE57338dataset, including 2817 upregulated and 2892downregulated genes. Four LRG (JMJD1C, EIF4G1, PPIL4 and PHC3) were identified as feature genes related to ICM using two machine learning algorithms. The diagnostic model constructed with these feature genes can accurately distinguish ICM, with decision and calibration curves demonstrating robust performance. External validation set confirmed high diagnostic accuracy, and the area under the curve (AUC) is 0.909. The results of gene set enrichment analysis (GSEA) revealed that immune-inflammatory responses and metabolic dysregulation may play critical roles in ICM pathogenesis. Conclusion This study successfully identified four LRG as potential biomarkers for ICM, and established a novel ICM diagnostic model, which can accurately distinguish ICM patients, and provide new insights for precision diagnosis and treatment. |
| keywords:Ischemic cardiomyopathy Lactylation-related gene Machine learning Bioinformatics Diagnostic model |
| 查看全文 查看/发表评论 下载PDF阅读器 |
|
|
|
