Evolutionary and clinical neocentromeres: two faces of the same coin?