Breast cancer is a highly heterogeneous disease; however, high data content assays have helped delineate the complexity and diversity of breast cancer. The Cancer Genome Atlas (TCGA) is characterizing over 1000 breast cancers as part of our Research Network. For most samples, there is data available for exome sequencing, SNP microarrays for copy number and genotypes, mRNA-seq, miRNA-seq, DNA methylation arrays, and reverse phase protein arrays for proteomic analysis representing one of the largest data sets of breast cancers. An integrative clustering analysis using individual subtypes defined from each of five platforms confirmed the presence of the previously-defined intrinsic subtypes in breast cancer: Basal-like, HER2-enriched, Luminal A and Luminal B. We identified patterns of expression, mutation, copy number aberrations, and DNA methylation that associated with each of these subtypes. Importantly, not only is the Basal-like subtype the most distinct subtype within breast cancer, but an analysis of 3,500 tumors from 12 different cancers found that the Basal-like tumors were in a class by themselves. In addition, they were more similar to ovarian and squamous samples of the lung than to other tumors originating from the breast. Whole exome sequencing of almost 1000 tumors now identifies four significantly mutated genes at greater than 10% in breast cancer including TP53 and PIK3CA (34% each), CDH1 (13%) and GATA3 (12%). However, their mutation frequencies varied widely by subtype (for example, TP53 is mutated in 11% of Luminal A and 88% of Basal-like). We identified both subtype-specific mutated genes as well as genes with differences in their mutation spectrum according to tumor subtype. Integrative pathway analysis has found several pathways with recurrent alterations in most tumors. In particular, the PI(3)K/Akt signaling pathway was significant altered and showed differences between basal and non-basal subtypes. Continued integrative analysis will help us develop a better understanding of events that occur in breast cancer and identify drivers for each of the unique subtypes.