基于生物信息学的胃癌特征基因网络关键节点及预后关联分析

目的:本研究旨在利用生物信息学筛选胃癌的特征基因，探索胃癌的恶性机制。方法:通过GEO数据库筛选胃癌相关基因芯片作为训练集，并利用GEO2R工具分析胃癌组织相较于正常组织的差异基因，进而在验证集患者中对所筛选基因进行验证；采用String数据库计算其蛋白质-蛋白质相互作用(protein-protein interaction,PPI)网络。根据Cytoscape软件的Centiscape等插件分析PPI网络中的关键节点，分析蛋白相互作用相关性；并利用DAVID数据库对差异基因和PPI关键模块基因进行基因功能富集及注释。随后对关键节点基因进行生存分析。结果:所筛选出的63个特征差异基因对胃癌与正常组织区分度良好，主要参与调控细胞外基质受体相互作用、PI3K-AKT信号通路等。PPI网络中关键节点调控肿瘤增殖、转移。对关键节点基因进行生存性分析发现，ITGB1、COL1A2表达高的患者，其生存率远低于低表达患者（P<0.05）。结论:本文可为寻找胃癌发生发展的关键基因，探索胃癌治疗新靶点提供一定的理论依据。

A correlation analysis between survival rate and the characteristic gene of gastric cancer based on bioinformatics

AIM: To investigate the possible mechanism of gastric cancer by analyzing the differences of gene modules and key pathways in gastric cancer patients, then look for effective treatment based on the feature genes. METHODS: Gene expression profiles of the gastric cancer in GEO database were selected. We used GEO2R tools to identify differential expression genes (DEGs) and String database was employed to conduct visualization analysis for protein-protein interaction (PPI) network.Then, the PPI network was imported into the Cytoscape software to find key nodes.After that, we employed the DAVID database to enrich and annotate the pathway and the interactions with key modules.RESULTS:Our study found 63 characteristic genes of gastric cancer, involved in regulation of extracellular matrix receptor interaction and PI3K-AKT signal pathway. ITGB1, COL1A2 were key nodal proteins which related to tumor proliferation and metastasis, and their expression were strongly associated with poor survival (P<0.05). CONCLUSION: Our study employs bioinformatics method from various perspectives to define the gene expression characteristics of gastric cancer which will provide a theoretical basis for the new target of gastric cancer.