Cancer is a disease in which abnormal cells acquire the ability to divide without control and invade other tissues. These attributes arise due to genetic and epigenetic alterations that disrupt molecular pathways regulating genomic stability, cell proliferation, and cell death. Since, cancer data analysis involves large numbers of variables, high dimensional clustering methods are required to interpret the data. High dimensional clustering is an efficient computational method for grouping similar objects into groups by understanding their large set of attributes simultaneously. Proposed research focuses on overcoming the pitfalls in existing tools. For developing molecular signatures to differentiate tumors, several computational algorithms should be integrated and the output should be validated against biological pathways data. In this study, molecular features which discriminate between tumor types will be annotated according to their role in cancer-critical processes to provide clearer biological and clinical insights.