生物信息学分析胰腺癌组织关键基因表达及其意义

Expression and Significance of Key Genes in Pancreatic Cancer Tissues:A Bioinformatics Analysis

目的利用生物信息学方法对GEO数据库进行探索,获取胰腺癌发生、发展的核心基因,分析胰腺癌发生、发展的潜在机制。方法使用GEOquery包从GEO数据库获取胰腺癌相关芯片数据(GSE62452),利用Limma包进行差异分析,结合clustreProfiler包对上调基因和下调基因分别进行GO功能和KEGG通路分析,利用String在线网站对差异基因进行蛋白互作网络分析,利用Cystoscopes软件筛选出核心基因,最后借助TCGA数据库和芯片数据(GSE28735)对核心基因进行再次验证。结果利用生物信息学方法共获得286个差异基因,其中包含上调基因189个,下调基因97个。富集分析结果显示上调基因涉及细胞组织黏附、细胞外基质组织构成等功能以及蛋白消化与吸收、PI_(3)K-Akt等信号通路相关。下调基因与脂类消化、细胞壁破裂等功能和胰腺分泌等信号通路密切相关。通过蛋白互作网络筛选出20个关键基因,利用GEPIA数据库对关键基因进行生存分析验证,发现3个基因(MMP11、LAMB3、PLAU)与胰腺癌预后明显相关。结论通过生物信息学方法最终筛选出3个核心基因,为胰腺癌的诊断和预后提供了新的思路。

Objective Gene expression omnibus(GEO)databases was analyzed by using bioinformatics to screen the key genes,and to analyze the potential mechanisms of pancreatic cancer.Methods Microarray data of GSE62452 were obtained from GEO databases using GEOquery package and differentially expressed genes were screened using Limma package.With the clustreProfiler package,GO function and KEGG pathway analysis were carried out for up-regulated genes and down-regulated genes respectively.Protein interaction network analysis of differentially expressed genes was analyzed using String online websites,and Cystoscopes software was used to screen core genes.Finally,the core genes were verified with TCGA database and chip data(GSE28735).Results A total of 286 differentially expressed genes were extracted by bioinformatics,including 189 up-regulated genes and 97 down-regulated genes.Enrichment analysis revealed that upregulated genes were closely associated with the function and pathway enrichment such as cellular tissue adhesion,extracellular matrix tissue composition,protein digestion and absorption,and PI_(3)K-Akt signaling pathways.Down-regulated genes are closely related to lipid digestion,cell wall rupture and pancreatic secretion signaling pathways.Twenty core genes were screened using the protein interaction network,and survival analysis of key genes were performed using the GEPIA database.We found that three genes(MMP11,LAMB3,PLAU)were significantly correlated with the prognosis of pancreatic cancer.Conclusion Three core genes were selected by bioinformatics method,which provided a new idea for the diagnosis and prognosis of pancreatic cancer.