Figure 2From: A new unsupervised gene clustering algorithm based on the integration of biological knowledge into expression dataMatrix T coexp : decomposition of the matrix T . Decomposing biological functions into coexpressed biological functions leads to build the matrix T c o e x p where a row represents a gene and a column a coexpressed biological function. The general term of T c o e x p , T kj 1 k ∈ K l j equals 1 if the gene k is associated with the function j and if it belongs to the cluster K l j , 0 else wise. The column margin of the coexpressed biological function l is equal to the number of genes in the corresponding cluster, that is card( K l j ). In addition, for every function j, the sum of the column margins associated with the coexpressed biological functions derived from j is equal to the column margin associated with the function j: ∑ l = 1 L j card( K l j )= T .j . Finally, we can remark that the row margins and the total number of associations are equal to those from T.Back to article page