应用GSEA和WGCNA方法分析细菌性败血症宿主关键差异基因及意义

Bioinformatics analysis of hub genes and correlative pathways in the host pathogenesis of bacterial sepsis by gene set enrichment analysis and weighted gene co-expression network analysis

摘要:

[目的] 采用生物信息学方法分析公共数据库来源的细菌性败血症患者全血转录组学表达谱，探讨细菌败血症相关的宿主关键差异基因及意义。[方法] 基于GEO数据库中GSE80496和GSE72829全血转录组基因数据集，采用GEO2R、基因集富集分析（GSEA）联用加权基因共表达网络分析（WGCNA）筛选细菌性败血症患者相比健康人群显著改变的差异基因，通过R软件对交集基因进行GO功能分析和KEGG富集分析。同时，通过String 11.0和Cytoscape分析枢纽基因，验证枢纽基因在数据集GSE72809（Health组52例，Defined sepsis组52例）全血标本中的表达情况，并探讨婴儿性别、月（胎）龄、出生体重、是否接触抗生素等因素与靶基因表达谱间的关系。[结果] 分析GSE80496和GSE72829数据集分别筛选得到932个基因和319个基因，联合WGCNA枢纽模块交集得到与细菌性败血症发病相关的10个枢纽基因（MMP9、ITGAM、CSTD、GAPDH、PGLYRP1、FOLR3、OSCAR、TLR5、IL1RN和TIMP1）；GSEA分析获得关键通路（氨基酸糖类-核糖代谢、PPAR信号通路、聚糖生物合成通路、自噬调控通路、补体、凝血因子级联反应、尼古丁和烟酰胺代谢、不饱和脂肪酸生物合成和阿尔兹海默症通路）及生物学过程（类固醇激素分泌、腺苷酸环化酶的激活、细胞外基质降解和金属离子运输）。[结论] 本项研究通过GEO2R、GSEA联用WGCNA分析，筛选出与细菌性败血症发病相关的2个枢纽模块、10个枢纽基因以及一些关键信号通路和生物学过程，可为后续深入研究细菌性败血症致病机制奠定理论依据。

Abstract:

[Objective] Bioinformatics methods were used to analyze the whole blood transcriptome data of patients diagnosed with bacterial sepsis from public databases, and we explored the hub genes related to bacterial sepsis and their clinical significance.[Methods] The datasets GSE80496 and GSE72829 were obtained from Gene Expression Omnibus (GEO). GEO2R, gene set enrichment analysis (GSEA) and weighted gene co-expression network analysis (WGCNA) were applied to screen out the differentially expressed genes (DEGs) among the patients with bacterial sepsis, when compared to the healthy. Gene ontology (GO) analysis and Kyoto Encyclopedia of Genes and Genome (KEGG) enrichment analysis were also performed on the key genes by R software, hubs genes were obtained by using String 11.0 and Cytoscape subsequently. In addition, the dataset GSE72809 was involved to verify the expression profiles of hub genes in whole blood samples among infants. And the relationships between gene expression and samples' baseline data (such as:gender, gestational age, birth weight, and antibiotic exposure) were also discussed in the study.[Results] 932 DEGs were screened out in GSE80496 dataset, while 319 DEGs were screened out in GSE72829 dataset. We identified 10 hub genes (MMP9, ITGAM, CSTD, GAPDH, PGLYRP1, FOLR3, OSCAR, TLR5, IL1RN and TIMP1). GSEA analysis shows key pathways (amino acid carbohydrate ribose metabolism, PPAR signaling pathway, glycan biosynthesis, autophagy regulatory pathway, complement/coagulation factor cascade reaction, nicotine and nicotinamide metabolism, unsaturated fatty acids biosynthesis and Alzheimer's disease pathway) and biological processes (steroid hormone secretion, adenylate cyclase activation, extracellular matrix degradation and metal ion transport) associated with bacterial sepsis.[Conclusion]] Through the analysis of GEO2R, GSEA combined with WGCNA, our study screened out 2 pivotal modules, 10 hub genes, several signaling pathways and biological processes closely related to bacterial sepsis, which may lay a theoretical basis for further research on the pathogenesis of bacterial sepsis.