一种新的基因注释语义相似度计算方法  

A new approach to measure the semantic similarity of gene annotation

作　　者：吴飞珍[1] 马文丽[1,2] 王旺迪[3] 陈启龙[1] 郑文岭[1,2] 

机构地区：[1]上海大学电子生物中心,上海200072 [2]南方医科大学基因研究所,广州510515 [3]浙江东方职业技术学院,浙江325011

出　　处：《生物信息学》2010年第1期23-29,共7页Chinese Journal of Bioinformatics

基　　金：上海大学研究生创新基因资助

摘　　要：基因本体(GO)数据库为基因提供了统一的注释,有效地解决了不同数据库描述相同基因的不一致问题。但是,根据基因注释如何比较基因的功能相似性,这个问题仍然没有得到有效解决。本文提出一种新的基因注释语义相似度计算方法,这种方法在本质上是基于基因的生物学特性,其特点在于结点的语义相似度与结点所在集合无关,只与结点在GO图的位置有关,语义相似度可被重复利用。它既考虑了基因所映射的GO结点深度,又考虑了两GO结点之间所有路径对结点语义相似度的影响。文中以酵母菌的异亮氨酸降解代谢通路和谷氨酸合成代谢通路为实验,实验结果表明这种算法能准确地计算基因注释语义相似度。Although the gene ontology（GO） provides a consistent gene annotation,which effectively tackles the problem that the same genes are described as different biological vocabularies in heterogeneous data sources.However,there is still no effective method to determine the functional similarities of genes based on gene annotation information.The paper proposed a new method to measure the semantic similarities of gene annotations.The approach,in essence,is based on biological property of genes.The semantic similarities of GO terms measured by the method are only relative to the characteristics of these GO terms,but not to the set that these GO terms belong to.The method considers not only the depth of GO terms,but also the contribution of every path distance between two GO terms.The approach has been applied to analysis of isoleucine degradation pathway and super-pathway of glutamate biosynthesis in Saccharomyces cerevisiae.The results show the method may exactly measure the semantic similarities of gene annotations.

关 键 词：基因本体 基因注释 语义相似度 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]