We sought to build a catalog of epitopes presented by breast cancers using a renewable resource of well-characterized breast cancer cell lines. Starting from 70 breast cancer cell lines, we measured MHC class I abundance and used pre-existing RNAseq data to identify either HLA-A*02 or MHC class I-positive cell lines. For 20 of these cell lines, we used “reverse” immunogenetics, in which MHC class I-loaded peptides are recovered and their sequences are determined by mass spectrometry. We identified more than 2,700 unique MHC class I-bound peptides from a panel of basal, luminal, and claudin-low subtype of cell lines. HLA-A*02 binding prediction across all tested cell lines revealed a model which described the distribution of HLA-A*02-binding peptides and allowed us to identify those peptides most likely to be presented on HLA-A*02. Comparing the peptides that we identified to published literature found that more than 1500 peptides had been identified in previous studies and that 18 of these peptides have been shown to be immunogenic. Overall, this high throughput identification of MHC class I-loaded peptides is an effective strategy for systematic characterization of cancer epitopes and could be employed in a design of multipeptide-based anticancer vaccine.