Each year in the United States, over 75,000 individuals are diagnosed with melanoma and over 9,500 individuals die from the disease. When detected early, melanoma is highly curable, with 5-year survival rates of 98% for early stage disease. However, the survival rate drops precipitously for later stages, declining to 62% for regional stages and 16% for distant stages. Advances in the ability to individuals at high riskof developing melanoma in conjunction with improvements in prevention, screening, and early detection have the potential to dramatically reduce disease mortality. In this project, we propose to conduct a hypothesis-free, genome-wide search for novel melanoma-susceptibility genes characterized by rare, protein-coding risk variants. This project builds upon the rich specimen repositories and comprehensive family, epidemiologic, and exposure data at MD Anderson and the University of Utah. There are three specific aims. In Aim 1, we will sequence the whole exomes of 1,200 melanoma cases and 1,200 frequency-matched controls. The cases will include 250 individuals from 125 three- to five-generation high-risk melanoma pedigrees recruited at the University of Utah and 950 familial index cases recruited at MD Anderson with at least one affected first-degree relative. We will perform gene-based tests to identify the top 1,000 candidate genes (p<0.05) using the pedigree Variant Annotation, Analysis & Search Tool (pVAAST), a state-of-the-art method that integrates familial linkage, case- control association, and variant prioritization information in a unified statistical framework. We will also use the offtarget reads from the exome sequence data to impute single nucleotide variants (SNVs) for each individual across the entire genome. In Aim 2, we will validate the 1,000 candidate genes and top 1,000 SNVs through targeted sequencing in 3,000 melanoma cases and 3,000 controls. Together, Aims 1 and 2 have high power identify novel, intermediate-risk melanoma-susceptibility genes across the entire genome. We will also integrate the genetic risk factors intoan existing quantitative melanoma risk prediction model using a subset of 2,000 cases and 1,200 controls for which comprehensive epidemiologic and exposure risk factor data is available. In Aim 3, we will sequence 600 tumors from individuals with damaging rare susceptibility variants identified from Aims 1-2 to test for somatic-germline interaction. The familial and case-control efforts proposed in Aims 1 and 2 are designed to comprehensively identify and then replicate rare genetic risk factors for melanoma on a genome-wide basis. The newly identified risk factors from this study will enable the identification of genetically susceptible individuals at high risk of developing melanoma. The incorporation of somatic variation information will provide additional insight into the role of germline susceptibility varints in cancer progression. The integration of epidemiologic and exposure risk factors will further enhance the translational potential of this work, leading to more effective prevention, early detection, and treatment strategies for this deadly yet often preventable disease.