用基因互作网络识别胰腺导管腺癌基因生物标记  

Identification of Gene Biomarkers for Distinguishing Pancreatic Ductal Adenocarcinoma from Normal Tissue Using a Network-based Approach

作　　者：王新格[1] 田卫东[1] 

机构地区：[1]复旦大学生命科学学院生物统计学与计算生物学系,上海200438

出　　处：《复旦学报（自然科学版）》2016年第5期642-648,共7页Journal of Fudan University：Natural Science

摘　　要：胰腺导管腺癌(PDAC)是全球高致死率癌症中的一种.PDAC基因生物标记的识别可以通过构建基因互作网络完成.利用蛋白质互作网络来分析与研究基因表达芯片数据,构建出PDAC基因互作网络并对其划分基因模块,进而筛选出在模块中的PDAC差异性表达基因.通过筛选在癌症样本和正常组织样本中共表达的基因对,并利用STRING蛋白质互作网络评估基因功能相关性,构建出具有PDAC特异性的基因互作网络.利用iNP算法进行网络模块化分,在每一个模块中,模块内基因都具有强的共表达特性和模块功能相关性.通过筛选,获得了34个基因模块,其中20个在癌症样本中表达明显上调,14个在癌症样本中表达明显下调.从这些模块中又筛选出在PDAC样本中表达上调的10个基因生物标记,如DMBT1、DSC3等和表达下调的10个基因生物标记,如DLG5、NRCAM等.Pancreatic ductal adenocarcinoma（PDAC） is one of the most lethal cancers in the world. In this study, we have developed a network-based approach to identify gene biomarkers that can distinguish PDAC and normal tissues. By identifying strongly coexpression gene pairs in normal tissues and PDAC samples and applying functional association information from STRING network online, a PDAC specific gene association network was constructed. Based on this network, gene modules was obtained in which genes are highly functionally associated using iNP algorithm. Then gene modules which are differentially expressed between PDAC and normal samples were identified. Finally, genes inside those modules were selected with discriminating coexpression patterns between PDAC and normal samples and those genes can be used as candidate biomarkers to facilitate clinicopathologic diagnose.

关 键 词：胰腺导管腺癌 基因生物标记 基因互作网络 功能富集分析 

分 类 号：Q332[生物学—遗传学]