铁死亡相关基因在增殖型狼疮肾炎中的生物信息学研究  

Identification of ferroptosis-related genes in proliferative lupus nephritis by bioinformatics approach

作　　者：孔杰 高瑛瑛[1] 达展云[2] Kong Jie;Gao Yingying;Da Zhanyun(Department of Rheumatology,Nantong First People's Hospital,the Second Affiliated Hospital of Nantong University,Nantong 226000,China;Department of Rheumatology,the Affiliated Hospital of Nantong University,Nantong 226000,China)

机构地区：[1]南通市第一人民医院、南通大学第二附属医院风湿免疫科,南通226000 [2]南通大学附属医院风湿免疫科,南通226000

出　　处：《中华风湿病学杂志》2024年第7期460-464,I0007,I0008,共7页Chinese Journal of Rheumatology

基　　金：南通市科技局资助课题(MS12021053)。

摘　　要：目的通过公共基因芯片数据库(GEO),分析铁死亡相关基因(FRGs)在增殖型狼疮肾炎(PLN)中的表达及与临床指标的相关性。方法从GEO数据库中下载GSE65391数据集,运行R语言的limma包,采用Wilcox秩和检验,对514例PLN患者与72名健康对照者的外周血基因表达量进行差异分析。然后对差异基因进行基因本体论(GO)和京都基因与基因组百科全书(KEGG)分析。最后通过随机森林树(RF)、支持向量机(SVM)、有监督模型(XGB)和广义线性模型(GLM)4种机器学习算法进一步筛选核心基因,采用Spearman相关系数分析其与临床指标的相关性。结果PLN患者与健康对照组相比,有38个差异FRGs,其中16个上调,22个下调。GO富集显示差异基因的生物学过程集中在细胞对化学应激的反应,细胞组成主要为转录调节复合物,分子功能主要为结合DNA转录因子。KEGG通路主要为NOD样受体信号通路。GLM算法为最佳的机器学习算法,根据基因重要性评分筛选出10个基因作为核心基因,其中Myb原癌基因(MYB)的评分最高,其与SLEDAI评分(r=0.21,P<0.001)、ALT(r=0.20,P<0.001)、AST(r=0.18,P<0.001)、LDH(r=0.31,P<0.001)、肌酐(r=0.24,P<0.001)和ESR(r=0.22,P<0.001)等指标呈正相关,与白蛋白(r=-0.28,P<0.001)呈负相关。结论FRGs可能为研究PLN的潜在发病机制提供新的思路。Objective To analyze the differential expressions of ferroptosis related genes(FRGs)in proliferative lupus nephritis and its correlation with clinical indices(PLN)by analyzing the gene expression omnibus(GEO)database.Methods The GSE65391 dataset was downloaded from GEO database,differential FRGs of PLN were screened by limma package of R language.Wilcoxon rank-sum test was used to analyze the differentially expressed genes in peripheral blood between 514 PLN patients and 72 healthy controls.Gene ontology(GO)and Kyoto encyclopedia of genes and genomes(KEGG)enrichment analysis of differential FRGs were determined with the clusterProfiler package.Hub genes were determined using random forest(RF),support vector machine(SVM),xgboost(XGB)and generalized linear model(GLM)machine learning algorithms.Finally,Spearman rank correlation analysis was used to analyze the correlation between hub genes and clinical indices.Results A total of 38 differential FRGs of PLN patients were screened out,of which 16 were up-regulated and 22 were down-regulated.Biological process was enriched in cellular response to chemical stress,cellular component was enriched in transcription regulator complex,molecular function was enriched in DNA-binding transcription factor binding.KEGG enrichment analysis showed that FRGs were mainly involved in NOD-like receptor signaling pathway.GLM algorithm was selected to predict gene essentiality according to the area under the receiver operating characteristic(ROC)curve,10 hub genes were determined,of which MYB was the most important.MYB was positively correlated with SLEDAI(r=0.21,P<0.001),ALT(r=0.20,P<0.001),AST(r=0.18,P<0.001),LDH(r=0.31,P<0.001),Cr(r=0.24,P<0.001)and ESR(r=0.22,P<0.001)and negatively correlated with albumin(r=-0.28,P<0.001).Conclusion FRGs may provide new insight into the potential mechanisms of the pathogenesis of PLN.

关 键 词：红斑狼疮 系统性 狼疮肾炎 铁死亡 计算生物学 

分 类 号：R593.242[医药卫生—内科学]