Global-Local Dirichlet Processes for Clustering Grouped Data in the Presence of Group-Specific Idiosyncratic Variables

Motivation • Cancer is not a single disease • Classification of cancer -a step towards personalized treatment • Traditional classification of cancer restricted to tumor site of origin • Recent studies have shown different cancers have significant clinical and molecular similarities • Classification is done through genomic similarities • Clinical variables/prognostic biomarkers not used in the classification