Implementation χ–sim co-similarity and agglomerative hierarchical to cluster gene expression data of lymphoma by gene and condition

Abstract

Application chi-sim co-similarity and agglomerative hierarchical clustering in this study are used to clustering gene expression data of Lymphoma by gene and condition. The process begins by taking the gene expression data of Lymphoma, after that microarray gene expression data of Lymphoma will be standardized by using standardized rows and columns. Then, the concept of chi-sim co-similarity applied to create the matrix similarity row (SR) and similarity column (SC). The matrix elements of SR and SC are normalized by using a pseudo normalization. Finally, we use three approaches in agglomerative hierarchical clustering to cluster the data by gene and condition. Three approaches in agglomerative hierarchical clustering are single linkage, average linkage, and complete linkage. The result of clustering by column and gene in this study, give us the best outcome when complete linkage in agglomerative hierarchical clustering is combined with chi–sim co-similarity compared with single linkage and average linkage in agglomerative hierarchical clustering combined with chi-sim co-similarity.