Breast cancer genomes are known to undergo extensive epigenetic modification early on during tumorigenesis, including both DNA hyper- and hypomethylation. While hypermethylation has been extensively studied, detailed analysis of hypomethylation, which primarily occurs at repetitive DNA in cancer, has been difficult to study due to technological limitations. We are developing new genome-wide experimental and computational tools to understand hyper- and hypomethylation events in breast cancer and how these changes remodel the transcriptome. In particular, we are asking which regions in and near a gene promoter enforce transcriptional silencing when methylated and how hypomethylation of normally silent promoter sequences contained within repetitive DNA can activate nearby genes. We are examining how these changes correlate with clinical data, such as response to endocrine therapy and survival. In addition, we are using these results to suggest mechanistic links between hypomethylation and genomic instability, which are frequently correlated in breast cancer.