人类肾脏组织特异性蛋白网络构建及分析  

Interaction network construction and enrichment analysis for human kidney tissue specific proteins

作　　者：张小驰[1,2] 宫秀军[1,2] 

机构地区：[1]天津大学计算机科学与技术学院,天津300072 [2]天津市认知计算与应用重点实验室,天津300072

出　　处：《广西大学学报（自然科学版）》2013年第5期1108-1116,共9页Journal of Guangxi University（Natural Science Edition）

基　　金：国家自然基金资助项目(61170177);国家973项目(2013CB32930X)

摘　　要：生命体内组织特异性基因往往在对应的组织中表现出高的共表达性,在组织特异性基因的调控中,非组织特异性的转录因子往往起很大作用,并且这些转录因子往往与特定的转录因子一起影响组织的特异性,因此研究组织特异性基因编码蛋白相互作用网络必须考虑非组织特异性蛋白的影响。本文提出了一种利用最短路径算法来计算组织特异性基因编码蛋白的关联蛋白,从而构建最大连接强度的组织特异性蛋白相互作用网络,并对其拓扑结构进行基因本体(GO)、KEGG Pathway和疾病本体(DO)的富集度分析。通过对肾脏组织中的1 486个蛋白质及其相应的4 011条蛋白质相互作用分析,发现绝大部分结构的功能与肾脏组织的功相吻合,同时也发现了几种比较有趣的表面上与肾脏组织无关的功能及疾病。In the living body, tissue-specific （TS） genes often exhibit high co-expression in the cor- responding tissue. While in the regulation of the TS genes, the transcription factors （TF） of non TS genes often play a significant role, and these TFs along with specific TFs contribute to the functional- ity of tissue specificity. Therefore, it is necessary to take into account of these non TS proteins when studying the interaction networks of proteins encoded by TS genes. A method is presented to build TS protein interaction networks incorporating non TS proteins with high connectivity by using the shortest path algorithm in this paper. Furthermore, enrichment analysis is performed using three knowledgebase （ Gene Ontology ： GO ; KEGG Pathway ; and disease ontology ： DO） on key compo- nents of above networks. From the analysis of 1 486 proteins and 4,011 corresponding protein inter- actions in the kidney tissue, it is found that functions of most clusters are consistent with their tissue origins. Meanwhile some interesting functionality and diseases that are not linked to kindey function are found.

关 键 词：组织特异性 蛋白质相互作用网络 肾脏 最短路径 富集度分析 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]