【摘要】 为了挖掘基因表达数据中的差异共表达致病基因模块,提出了基于互信息和最大团相结合的方法.互信息用于度量基因表达谱之间的相互关系,计算任意2条基因表达谱在2种不同样本中的互信息值,得到2个互信息矩阵M1和M2,选定2个阈值T1和T2(T1>T2)将矩阵M1和M2二值化,并通过M1和M2中元素的逻辑"与"运算得到图的邻接矩阵,从邻接矩阵挖掘出的最大团则为差异共表达致病基因模块.将该方法应用于Colon数据,选定T1=2.2,T2=1.0,得到6个相互重叠的最大团,实验结果表明,该方法能有效挖掘出差异共表达致病基因模块.更多还原

【Abstract】 Mutual information combined with clique analysis method is introduced to identify differentially co-expressed disease-related genes at the level of biological modules from gene expression data.The mutual information is used to measure the co-expression relationships of gene data.Two square symmetric mutual information matrices M1and M2 are obtained by calculating the values of mutual information between each pair of genes in two different kinds of samples.Threshold values T1 and T2(T1>T2) are chosen respectively for the binarization of M1 and M2.The adjacency matrix of graph is obtained by logical operation "AND" of M1 and M2.The cliques detected from the adjacency matrix represent the differentially co-expressed disease-related gene modules.This method was applied to Colon microarray data with T1=2.2,T2=1.0,six overlapped cliques were detected.The results indicate that this method is very efficient in identifying differentially co-expressed disease-related gene modules from gene expression data.更多还原