Breast cancer is one of the most common cancers in the US with approximately 227,000 new cases of invasive breast cancer and 40,000 breast cancer deaths predicted in 2012. Breast cancer has a strong heritable component with approximately 15% to 20% of cases exhibiting a family history of the disease. Susceptibility to breast cancer is associated with rare germline variants in high-risk genes such as BRCA1 and BRCA2, several intermediate-risk (3 to 5 fold) predisposition genes such as PALB2 and CHEK2, and many common genetic variants associated with modest (< 1.5 fold) increased risk of disease. Currently, high-risk genes and intermediate risk genes are used for clinical genetic testing for breast cancer susceptibility and for clinical management of individuals with a family history of breast cancer. However, the known predisposing variants account for less than 50% of all familial breast cancer cases. Thus, many individuals with a family history of breast cancer cannot benefit from informative clinical genetic testing and enhanced cancer risk assessment and management. Although non-genetic factors and additional common genetic variants also may influence breast cancer risk, it is unlikely that these additional factors account for all of te missing heritability of breast cancer. Thus, we hypothesize that a significant amount of the unexplained familial risk of breast cancer is due to rare genetic variants that are associated with intermediate-to-high risk. Herein, we propose to identify and characterize novel breast cancer susceptibility genes using a comprehensive sequence-based approach. We have already completed whole exome sequencing of multiple germline DNA samples from 200 high-risk breast cancer families and now propose to leverage the results from these exome sequencing studies to establish the contribution of candidate variants and genes to breast cancer. In Aim 1, we will validate 400 candidate genes in a case-control study of 4,000 familial breast cancer cases and 4,000 unaffected controls. In Aim 2 we will take a different approach to the identification of breast cancer risk factors by evaluating associations between rare recurring protein-coding variants and breast cancer risk. We will use a large case-control study of 8,000 breast cancer cases and 8,000 matched unaffected controls to validate candidates. Finally, in Aim 3 we will conduct functional studies of the candidate genes and variants from Aims 1 and 2 in order to improve prediction of pathogenic and non-pathogenic variants for the validation studies and to understand the signaling mechanisms associated with predisposition to breast cancer. The research team involved in this project has access to large, well annotated patient resources, has an established background in this research, is leveraging extensive preliminary data, and has the ability to utilize the findings for the benefit of breast cancer patients. Thus, his team is well positioned to account for much of the "missing heritability" of breast cancer.