胰腺导管腺癌差异基因表达及临床预后分析

Differential Gene Expression and Clinical Prognosis of Pancreatic Ductal Adenocarcinoma

目的分析胰腺导管腺癌(PDAC)差异基因表达及临床预后。方法从GEO(Gene Expression Omnibus)数据库中获取基因及微RNA(miRNA)数据集,并采用R 3.6.1分析数据集中的差异表达基因。采用注释、可视化和集成发现数据库(DAVID)对差异基因的基因本体论(GO)富集通路和京都基因与基因组百科全书(KEGG)通路进行分析,使用miRwalk在线数据库预测miRNA靶基因。通过STRING数据库、miRNet网站和Cytoscape软件构建蛋白质相互作用(PPI)网络和miRNA-基因网络。通过Kaplan-Meier plot数据库绘制生存曲线,并使用加利福尼亚大学圣克鲁斯分校Xena数据库构建基因表达分析聚类图。结果从GEO数据库4个基因数据集获得217个差异基因,2个miRNA数据集获得13个差异表达的miRNA,DAVID数据库中发现217个基因高度富集于8个生物学过程,8个细胞成分通路,5个分子功能通路及4个KEGG通路[伪发现率(FDR)<0.05]。连接系数>0.4的基因被用来构建PPI网络,共有167个基因通过分子复合物检测工具分析出10个重要模块。通过miRWalk数据库预测13个miRNA作用于6711个靶基因,将这些miRNA导入miRNet数据库中,发现4种miRNA在胰腺癌中作用于558个靶基因,最终得到了10个miRNA靶基因在PDAC中差异表达并高度富集于生物代谢途径。生存分析结果显示,纤连蛋白Ⅰ(FN1)、血小板反应蛋白2型(THBS2)、Ⅻ型胶原蛋白α-1链(COL12A1)、Ⅵ型胶原蛋白α-3链(COL6A3)和载脂蛋白L1型(APOL1)高表达组患者的生存率较低,而镁依赖性蛋白磷酸酶1δ(PPM1D)高表达组患者的生存率较高(P<0.05)。结论FN1、THBS2、COL12A1、COL6A3、APOL1和PPM1D的表达与PDAC患者的预后显著相关,有望成为潜在的治疗靶点。

Objective To analyze pancreatic ductal adenocarcinoma(PDAC)differential gene expressions and the clinical prognosis.Methods All datasets were obtained from the Gene Expression Omnibus(GEO)database analyzed by R 3.6.1 software.The Database for Annotation,Visualization and Integrated Discovery(DAVID)was used to analyze the gene ontology(GO)and Kyoto encyclopedia of genes and genomes(KEGG)pathway.miRwalk database was used for finding correlated genes of microRNA(miRNA).STRING database,miRNet website and Cytoscape software were executed to structure the protein-protein interaction(PPI)network and miRNA-gene network.The survival curves of identified hub genes were drawn by Kaplan-Meier plot database,the gene expression analysis cluster graph of which was constructed using University of California Santa Cruz Xena database.Results Altogether 217 differentially expressed genes(DEGs)were obtained from 4 datasets and 13 miRNAs were obtained from 2 miRNA datasets in GEO database.We found the 217 DEGs enriched in 8 biological processes,8 significant cellular components,5 molecular functions,and 4 important KEGG processe[false discovery rate(FDR)<0.05].DEGs with link coefficient>0.4 were used to construct the PPI networks,and a total of 167 genes were analyzed by molecular complex detection tools,and 10 important modules were identified.Totally 6711 genes were predicted related with 13 miRNAs in miRwalk database as target genes,and these miRNAs were introduced into miRNet database:4 miRNAs were found to act on 558 target genes in pancreatic cancer,finally,10 miRNA target genes differentially expressed in PDAC and highly enriched in the biological metabolic pathway were obtained.Survival analysis suggested that the survival rate of patients in the high expression group of fibronectin 1(FN1),thrombospondin 2(THBS2),collagen typeⅫalpha 1 chain(COL12A1),collagen typeⅥα-3 chain(COL6A3)was lower,while the survival rate of patients in the high expression group of protein phosphatase magnesium-dependent 1δ(PPM1D)was higher(P<0.05).Conclusion The expression of FN1,THBS2,COL12A1,COL6A3,APOL1 and PPM1D is highly correlated with the prognosis of patients with PDAC,which are expected to become potential therapeutic targets.