Evaluation of Order Preserving Triclustering (OPTricluster) in 3 Dimensional Gene Expression Data Analysis Using Gene Ontology

Breast cancer is the most common cancer in woman and very dangerous that caused 30% of all new female cancers for each year. Nowadays, bioinformatics can be usefull to analyze some disease, especially breast cancer. Triclustering is an improvement in bioinformatics and can be used for 3-dimensional (3D) data. Gene-sample-time is commonly used for 3-dimensional gene expression data. One of the Triclustering methods is Order Preserving Triclustering (OPTricluster). OPTricluster forms a tricluster by identifying genes that have a similar pattern of expression change across time points under several experimental conditions. In this study, OPtricluster was used to finding tricluster of breast cancer gene expression data, by observing changes in gene expression levels in several time conditions. After clustering, the genes which contained in each cluster were evaluated by Gene Ontology to find out which genes were affected by estrogen stimulation at each time point. The result is there were 77 triclusters with 3 triclusters of constant patterns, 52 triclusters of conserved patterns, and 18 triclusters of divergent patterns. For the constant, conserved, and divergent patterns, we found there are several genes that are most affected by estrogen stimulation i.e., NOP16, PTBP1, and ITGB4, respectively.