Medulloblastomas are the most common malignant brain tumors of childhood, with 40-50% overall mortality. My lab is focused on discovering the normal cells in the developing brain from which these tumors arise and the molecular mechanisms giving rise to the cancer, with a goal to develop more effective and less toxic targeted therapies.  Using genomic methods, we have found that certain subtypes of medulloblastomas arise from cerebellar granule cell progenitors. Using a Ptc +/- mouse model, we discovered that tumors arise through a multistep process, starting with persistent proliferation of pre-neoplastic cells followed by emergence of neoplastic cells that progress to form tumors. We now are testing how epigenetic mechanisms promote neoplastic transformation and tumor progression.