基于生物信息学和机器学习的克罗恩病关键基因筛选和免疫浸润分析  

Key Gene Screening and Immune Infiltration Analysis of Crohn's Disease Based on Bioinformatics and Machine Learning

作　　者：卜凡靖 BU Fanjing(Department of Gastroenterology,Binzhou Second People′s Hospital,Binzhou 256800,Shandong,China)

机构地区：[1]滨州市第二人民医院消化内科,山东滨州256800

出　　处：《医学信息》2025年第4期10-15,共6页Journal of Medical Information

摘　　要：目的利用生物信息学和机器学习算法筛选克罗恩病(CD)关键基因,并进行免疫浸润分析。方法下载基因表达数据库中包含CD和健康对照者(Hcon)的乙状结肠组织转录组测序数据进行差异分析,利用加权基因共表达网络分析(WGCNA)筛选CD相关的差异表达基因(DEGs)。利用最小绝对收缩和选择算子(LASSO)和随机森林(RF)等机器学习方法筛选CD的关键基因,并进行免疫浸润分析。结果共获得54个CD相关DEGs,机器学习算法筛选出CD的潜在生物标志物CCAAT增强子结合蛋白δ(CEBPD)。CD样本中静息树突状细胞比例低于Hcon样本。CEBPD与中性粒细胞呈正相关,与静息CD4记忆T细胞呈负相关。结论CEBPB是CD发病的关键基因,树突状细胞、中性粒细胞和CD4记忆T细胞与CD的发生发展密切相关,可能是治疗CD的关键途径。Objective To screen the key genes of Crohn′s disease(CD)by bioinformatics and machine learning algorithms,and to analyze the immune infiltration.Methods The transcriptional sequencing data of sigmoid colon tissue containing CD and healthy controls(Hcon)was downloaded for differential analysis,and weighted gene co-expression network analysis(WGCNA)was used to filter for CD-related differentially expressed genes(DEGs).The key genes for CD were identified using machine learning methods such as the least absolute shrinkage and selection operator(LASSO)and random forest(RF),and immune infiltration analysis was performed.Results A total of 54 CD-related DEGs were obtained,and machine learning algorithms identified the potential biomarker for CD,CCAAT/enhancer-binding protein delta(CEBPD).In CD samples,the proportion of resting dendritic cells was lower than in Hcon samples.CEBPD was positively correlated with neutrophils and negatively correlated with resting CD4 memory T cells.Conclusion CEBPB is a key gene in the pathogenesis of CD,and dendritic cells,neutrophils,and CD4 memory T cells are closely related to the development of CD,which may be a key in treating CD.

关 键 词：克罗恩病 免疫浸润 机器学习 生物信息学 

分 类 号：R574.4[医药卫生—消化系统]