Ben Berman University of Southern California (USC) Whole-genome bisulfite sequencing (WGBS) can be used to construct a comprehensive map of the epigenetic states in a patient tumor sample. While WGBS is a popular technique, current computational analysis methods fail to exploit its unique single-molecule properties. We have developed a new method, single-molecule epistate analysis, which takes as input two or more tumor WGBS samples, and uses these single-molecule patterns to identify epigenetic state changes genome-wide. Because the algorithm operates at the level of individual DNA molecule, it can be used to identify epigenetic state changes occurring within specific subpopulations of cells within a tumor sample. We will demonstrate results of this method using sequence data we have produced for dozens of tumors as part of NIH’s The Cancer Genome Atlas (TCGA) project.