Lymphoma is cancer of the lymphatic system that includes spleen, thymus, lymph nodes, bone marrow, and other parts of the body caused by lymphocytes cells. This study aims to implement the co-similarity measure in obtaining a non-overlapping significant bicluster in lymphoma gene expression data. Our proposed technique is based on χ -sim co-similarity measure for clustering row (gene) and column (condition). The process of biclustering is preceded by pre-processing using min-max normalization, applied co-similarity measure, partition each similarity matrix by using a k-means algorithm and extract bicluster. Furthermore, the silhouette method is used to determine the number of optimum clusters and measure the accuracy of grouping results. In our results, we obtained 2 clusters on the row (gene) similarity matrix and column (condition) similarity matrix with respectively silhouette value are 0.5464 and 0.7382. The clustering result consisted of 1242 genes (cluster-1) and 2784 genes (cluster-2) on the row (gene) similarity matrix and 41 conditions (cluster-1) and 21 conditions (cluster-2) on the column (condition) similarity matrix. Based on gene analysis that discovered genes in biclusters are supported by biological evidence and knowing the biological function of genes that significantly influence through gene ontology (GO) enrichment.