动脉粥样硬化组织中衰老基因的鉴定及免疫浸润分析

Identification of Aging Genes and Analysis of Immune Infiltration in Atherosclerosis

摘要: 目的：基于生物信息学方法来识别参与AS中的衰老基因。为靶向衰老细胞延缓衰老相关心血管疾病的发生提供理论依据。方法：我们在基因表达数据库(GEO)中选取GSE21545和GSE43292作为训练集，选取其中部分数据进行合并，去除批次效应。计算合并数据集的差异表达基因(DEGs)，并进行GO、KEGG富集分析。DEGs与衰老基因集(ARGs)取交集，得出筛选差异表达的衰老相关基因(DE-ARGs)，进行蛋白相互作用网络分析，分别采用MCC、MNC、EPC、Degree、Closeness、Radiality六种方法，选取各自前10位的Node基因取交集作为hub基因。同时对动脉粥样硬化组织的免疫细胞浸润情况进行分析，并构建hub基因调控的ceRNA网络。最后我们在GSE100927数据集中验证了hub基因的表达，以证实我们研究的普遍性。结果：我们从De-ARGs中确定了5个hub基因。功能富集分析表明，这些标记基因对氧化应激的反应，蛋白磷酸酶结合，PI3K-Akt信号通路等至关重要。通过免疫浸润分析发现，多种免疫细胞参与AS，hub基因与免疫细胞存在明显相关性。ceRNA调控网络显示，mRNA与miRNA、lncRNA之间具有一对多和多对一的复杂关系。利用外部数据集GSE100927，验证发现其中4个hub基因(IGF1, GRB2, MYC, PTEN)表达存在与训练集一致的显著性差异。结论：通过生物信息学方法，可以发现衰老基因在动脉粥样硬化组织中的表达情况及相关生物学功能，为靶向衰老细胞延缓衰老相关心血管疾病的发生提供理论依据。

Abstract: Objective: Identify aging genes involved in AS based on bioinformatics methods. It provides a theoretical basis for targeting aging cells to delay the occurrence of aging related cardiovascular diseases. Methods: We selected GSE21545 and GSE43292 as training sets in the Gene Expression Database (GEO), and merged some of the data to remove batch effects. Calculate differentially expressed genes (DEGs) from the merged dataset and perform GO and KEGG enrichment analysis. The intersection of DEGs and aging gene sets (ARGs) was obtained to screen differentially expressed aging related genes (DE-ARGs). Protein interaction network analysis was performed using six methods: MCC, MNC, EPC, Degree, Closeness, and Radiality. The intersection of the top 10 node genes was selected as the hub gene. At the same time, the infiltration of immune cells in Atherosclerosis tissue was analyzed, and the ceRNA network regulated by hub gene was constructed. Finally, we validated the expression of hub gene in the GSE100927 dataset to confirm the universality of our study. Results: We identified 5 hub genes from De-ARGs. Functional enrichment analysis indicates that these marker genes are crucial for their response to oxidative stress, protein phosphatase binding, and PI3K Akt signaling pathway. Through immune infiltration analysis, it was found that multiple immune cells are involved in AS, and there is a significant correlation between hub genes and immune cells. The ceRNA regulatory network shows a complex one-to-many and many-to-one relationship between mRNA, miRNA, and lncRNA. Using the external dataset GSE100927, it was found that there were significant differences in the expression of four hub genes (IGF1, GRB2, MYC, PTEN) consistent with the training set. Conclusion: Through bioinformatics methods, we can find the expression of aging genes in Atherosclerosis tissues and related biological functions, which provides a theoretical basis for targeting aging cells to delay the occurrence of aging related cardiovascular diseases.