Melanoma formation involves random mutational events with positive evolutionary selection acting on some of these events to produce driver mutations. Looking for these driver mutations in DNA initially focused on protein-coding somatic mutations. However, the availability of whole genome analysis has broadened the scope of enquiry to include potential driver mutations in non-coding regions of DNA. Mutations are abundant in this non-coding space, 50 times larger than the coding exome. Despite the number of events and size of the space, non-coding driver events appear relatively infrequently, and this is complicated by the difficulties of identifying signals of positive selection. Although rare, certain mutations occur outside the commonly targeted mutations and are more frequent in specific melanoma subsets, potentially offering therapeutic value. Beyond therapeutic application, these investigations increase our understanding of the mechanisms and complexity of DNA function and disruptive effects of mutation.