Tuberculosis (TB) is a disease with contagious that caused by various mycobacteria. Around 30% of our population is approximate to be infected with TB. Tuberculosis is still a deadly disease. Nowadays, there has been a lot of research on gene expression analysis to gain information about the diagnosis and prognosis of various diseases including TB. Clustering has been applied to gaining information or insight from a data distribution. In bioinformatics, clustering is used to learn about phenomenological responses of genes, predicted disease progression, and find biomarkers from a dataset in form of microarray data. Nowadays, it's possible to get the expression level of genes in a biological sample during a series of time points. Dataset with three-dimensional can be considered for gene expression subjects by clustering methods published. Triclustering analysis is a method on 3D data (observation, attribute, and context). Triclustering analysis can simultaneously group measurement on several attributes and contexts with an output subset of the data called tricluster. one of the triclustering method is THD-Tricluster that has advantages to form a tricluster from a data with shifting-and-scalling pattern. The results of the tricluster analysis can help diagnose various diseases as early as possible including TB.