基于生物信息学筛选胰腺导管腺癌关键基因

目的:利用生物信息学方法分析胰腺导管腺癌（PDAC）基因表达谱芯片并筛选关键基因。方法:从公共数据库基因表达数据库（GEO）中下载PDAC基因表达谱芯片GSE28735、GSE15471、GSE101448,共纳入108例PDAC样本和97例癌旁组织样本。应用R语言limma包和impute包筛选差异表达基因。利用DAVID数据库和在线分析工具Kobas分别对差异基因进行GO功能富集分析和KEGG通路富集分析。利用STRING数据库和Cytoscape软件构建差异蛋白互作网络并进一步筛选关键基因。结果:3个基因表达谱芯片共有161个差异表达基因（|log2 fold-change(FC)|>2，P<0.05），包括54个上调基因，107个下调基因。GO功能富集分析显示差异基因与extracellular exosome、extracellular space、extracellular matrix organization密切相关。KEGG通路分析显示差异基因主要富集在protein digestion and absorption、ECM-receptor interaction和focal adhesion等通路。蛋白质相互作用网络图中显示节点最多的10个枢纽基因分别是ALB、COL11A1、COL3A1、FN1、EGF、COL1A1、MMP9、COL5A2、ITGA2、COL6A3。结论:筛选所得的10个关键基因可能在PDAC发生发展中发挥重要作用，有望成为PDAC诊断及治疗的生物学靶标，为进一步研究PDAC发生发展的分子机制提供了理论依据。

Identification key genes of pancreatic ductal adenocarcinoma based on bioinformatics analysis

Objective:To explore key genes and pathways associated with pancreatic ductal adenocarcinoma(PDAC) by bioinformatics method for better understanding the underlying mechanisms.Methods:The gene expression profiles of GSE28735,GSE15471 and GSE101448 were obtained from Gene Expression Omnibus database,including 108 pancreatic ductal adenocarcinoma samples and 97 paracancerous tissue.DEGs were analyzed using R program by limma packages and impute packges.The online analysis tool DAVID and Kobas were used to perform the gene ontology(GO) and Kyoto Encyclopedia of Genes and Genomes(KEGG) pathway enrichment analysis.Protein-protein interaction network was constructed by STRING and visualized by Cytoscape software.The hub genes were identified by the Molecular Complex Detection(MCODE) plugin.Results:A total of 161 overlapped DEGs were identified,including 54 up-regulated genes and 107 down-regulated genes（|log2 fold-change(FC)|>2，P<0.05）.DEGs were significantly enriched in extracellular exosome,extracellular space,extracellular matrix organization.In addition,three KEGG pathway were significantly enriched,including pancreatic secretion,protein digestion and absorption,ECM-receptor interaction.The top ten hub genes,ALB,COL11A1,COL3A1,FN1,EGF,COL1A1,MMP9,COL5A2,ITGA2 and COL6A3 were identified from the PPI networks,and three significant modules as the sub-networks were detected,which were mainly involved in the protein digestion and absorption,ECM-receptor interaction and focal adhesion.Conclusion:The hub genes ALB,COL11A1,COL3A1,FN1,EGF,COL1A1,MMP9,COL5A2,ITGA2 and COL6A3 may be potential biomarkers and therapeutic targets for PDAC.Moreover,protein digestion and absorption,ECM-receptor interaction pathways play significant roles in the progression of PDAC.Our finding provides new insights into the pathogenesis of PDAC.