整合生物信息学法与加权基因共表达网络分析联合分析脑胶质瘤特征基因

Identification of Characteristic Genes in Glioblastoma by Integrated Bioinformatics and Weighted Gene Correlation Network Analysis

关键词:  脑胶质瘤  差异基因  生物信息学  加权基因共表达网络  转录因子

Key words:  glioblastoma  differentially expressed genes  bioinformatics  weighted gene correlation network  transcription factor

摘要:

目的 利用脑胶质瘤（glioblastoma，GBM）芯片数据，采用整合生物信息学法和加权基因共表达网络分析（weighted gene correlation network analysis，WGCNA）法，寻找肿瘤发病特征基因、关键通路以及转录调控机制。方法 利用GEO数据库中高通量基因芯片数据，通过整合生物信息学法筛选出差异基因；利用WGCNA分析GBM关键基因hub基因；采用维恩分析法，整合这些差异基因与hub基因，筛选出GBM特征基因。采用基因本体论（Gene Ontology，GO）基因功能注释，京都基因和基因组数据库（Kyoto Encyclopedia of Genes and Genomes，KEGG）通路富集，分析GBM特征基因所富集的功能和通路。利用Kaplan-Meier分析特征基因与GBM生成率的关系。利用基因云生物信息平台（GCBI），分析调控这些特征基因的转录因子。结果 经分析，发现273个特征基因。这些特征基因主要可影响离子通道、蛋白激酶、γ-氨基丁酸受体功能。CHD5，SYP，PHYHIP表达水平与GBM生存率显著相关；转录因子Sp1、Sp3、REST可能是调控这些特征基因的关键因子。结论 本研究从多种角度定义了GBM的特征基因及调控机制，为其精准治疗提供了依据。

Abstract:

OBJECTIVE To identify the key characteristic genes, pathways and transcriptional regulatory mechanisms of tumors using glioblastoma(GBM) chip data, integrated bioinformatics methods and weighted gene correlation network analysis (WGCNA). METHODS High throughput chip data was downloaded from GEO database. The integrated bioinformatics methods were used to identify the differentially expressed genes. WGCNA was used to analyze the hub genes, which was the key gene in GBM. Differentially expressed genes and hub genes were integrated by the Venn analysis, and the characteristic genes in GBM was identified. The Gene Ontology(GO) and Kyoto Encyclopedia of Genes and Genomes(KEGG) enrichment were used to analyze the functions and pathways enriched by GBM characteristic genes. Kaplan-Meier analysis was used to evaluate the association between the levels of the characteristic genes and patients' overall survival. The online tool Gene-cloud of Biotechnology Information(GCBI) was used to analyze the transcription factors regulating these characteristic genes. RESULTS Two hundred and seventy-three characteristic genes were identified. These genes most likely affected the functions of ion channel, protein kinase and GABA receptor. The expression level of CHD5, SYP, and PHYHIP were significantly related to the overall survival of GBM patients. Transcription factor Sp1, Sp3, and REST might be the key transcription factors for these characteristic genes. CONCLUSION The present study identifies the characteristic genes of GBM and their regulatory mechanism by various bioinformatics methods. The results may provide new basis for the precise treatment of GBM.