通过机器学习算法识别的炎症基因特征揭示了冠状动脉疾病的新型生物标志物。
Inflammatory Gene Signature Identified by Machine Algorithms Reveals Novel Biomarkers of Coronary Artery Disease.
作者信息
Liu Xing, Zhang Yuanyuan, Wang Yan, Xu Yanfeng, Xia Wenhao, Liu Ruiming, Xu Shiyue
机构信息
Department of Cardiology, The Third Affiliated Hospital, Sun Yat-sen University, Guangzhou, Guangdong, People's Republic of China.
Department of Hypertension and Vascular Disease, The First Affiliated Hospital, Sun Yat-sen University, Guangzhou, Guangdong, People's Republic of China.
出版信息
J Inflamm Res. 2025 Feb 10;18:2033-2044. doi: 10.2147/JIR.S496046. eCollection 2025.
PURPOSE
Inflammatory activation of immune cells plays a pivotal role in the development of coronary artery diseases (CAD). This study aims to investigate the immune responses of peripheral blood mononuclear cells (PBMCs) in CAD and to identify novel signature genes and biomarkers using machine learning algorithms.
METHODS
The GSE113079 dataset was analyzed and differentially expressed genes (DEGs) were identified between CAD and normal samples. The intersection of DEGs with inflammation-related genes was used to identify the differentially expressed inflammation-related genes (DIRGs). Then, the receiver operating characteristic (ROC) curves were plotted for each DIRG, and those with an area under the curve (AUC) greater than 0.9 were selected for subsequent analysis. Furthermore, machine learning algorithms were employed to identify biomarkers. A nomogram was developed based on these biomarkers. The CIBERSORT algorithm and Wilcoxon test method were used to analyze the differences in immune cells between the CAD and normal samples. The identified biomarkers were validated in PBMCs from patients with CAD and in atherosclerotic aortas from ApoE mice.
RESULTS
A total of 574 DEGs were identified between CAD and normal samples. From this intersection, 29 DIRGs were identified, of which 14 DIRGs (, , , , , , , , , , , , , and ) exhibited high diagnostic efficacy. Four biomarkers (, , , and ) were identified using Support Vector Machine (SVM). Ten types of immune cells, including CD8 T cells, regulatory T cells (Tregs), and resting NK cells, showed significant differences between the CAD and normal groups. Furthermore, increased levels of , , , and were validated in PBMCs isolated from CAD patients. In addition, , , and were upregulated in the mouse atherosclerotic aorta.
CONCLUSION
Our findings revealed novel inflammatory gene signatures of CAD that could be potential biomarkers for the early diagnosis of CAD.
目的
免疫细胞的炎症激活在冠状动脉疾病（CAD）的发展中起关键作用。本研究旨在调查CAD患者外周血单个核细胞（PBMC）的免疫反应，并使用机器学习算法识别新的特征基因和生物标志物。
方法
分析GSE113079数据集，确定CAD样本与正常样本之间的差异表达基因（DEG）。将DEG与炎症相关基因进行交集分析，以识别差异表达的炎症相关基因（DIRG）。然后，为每个DIRG绘制受试者工作特征（ROC）曲线，选择曲线下面积（AUC）大于0.9的DIRG进行后续分析。此外，采用机器学习算法识别生物标志物。基于这些生物标志物构建列线图。使用CIBERSORT算法和Wilcoxon检验方法分析CAD样本与正常样本之间免疫细胞的差异。在CAD患者的PBMC和ApoE小鼠的动脉粥样硬化主动脉中验证所识别的生物标志物。
结果
CAD样本与正常样本之间共鉴定出574个DEG。通过该交集分析，识别出29个DIRG，其中14个DIRG（……）具有较高的诊断效能。使用支持向量机（SVM）识别出4种生物标志物（……）。包括CD8 T细胞、调节性T细胞（Treg）和静息NK细胞在内的10种免疫细胞在CAD组和正常组之间存在显著差异。此外，在从CAD患者分离的PBMC中验证了……水平的升高。此外，……在小鼠动脉粥样硬化主动脉中上调。
结论
我们的研究结果揭示了CAD新的炎症基因特征，这些特征可能是CAD早期诊断的潜在生物标志物。
Zotero 插件