The Sequence Kernel Association Test (SKAT) is a statistical method designed to detect associations between rare and common genetic variants and complex traits. SKAT is a supervised, flexible, and computationally efficient regression approach that tests for association between genetic variants in a region and a continuous or dichotomous trait while adjusting for covariates. It uses a variance-component score test in a mixed-model framework to calculate p-values analytically, making it suitable for genome-wide data analysis. SKAT outperforms several alternative rare-variant association tests in simulations and real data analysis, including the Dallas Heart Study. It allows for the incorporation of flexible weights to enhance power, accounts for epistatic effects, and adjusts for covariates such as population stratification. SKAT is computationally efficient, requiring only fitting the null model to compute p-values. It is also robust to the direction and magnitude of effect, unlike burden tests which assume all variants affect the phenotype in the same direction. SKAT can be applied to both continuous and dichotomous phenotypes and is suitable for designing sequencing-based association studies. The method is validated through simulations and real data analysis, showing its effectiveness in detecting associations between rare variants and complex traits. SKAT provides analytic power and sample-size calculations to guide the design of new sequence association studies. The method is implemented in software and has been applied to real data, including the Dallas Heart Study, to test associations between rare variants and serum triglyceride levels. SKAT is the most powerful test for dichotomous traits and performs well for continuous traits compared to burden tests. The computation time for SKAT is efficient, making it suitable for genome-wide studies. SKAT is a flexible and powerful method for detecting associations between rare and common genetic variants and complex traits in sequencing-based association studies.The Sequence Kernel Association Test (SKAT) is a statistical method designed to detect associations between rare and common genetic variants and complex traits. SKAT is a supervised, flexible, and computationally efficient regression approach that tests for association between genetic variants in a region and a continuous or dichotomous trait while adjusting for covariates. It uses a variance-component score test in a mixed-model framework to calculate p-values analytically, making it suitable for genome-wide data analysis. SKAT outperforms several alternative rare-variant association tests in simulations and real data analysis, including the Dallas Heart Study. It allows for the incorporation of flexible weights to enhance power, accounts for epistatic effects, and adjusts for covariates such as population stratification. SKAT is computationally efficient, requiring only fitting the null model to compute p-values. It is also robust to the direction and magnitude of effect, unlike burden tests which assume all variants affect the phenotype in the same direction. SKAT can be applied to both continuous and dichotomous phenotypes and is suitable for designing sequencing-based association studies. The method is validated through simulations and real data analysis, showing its effectiveness in detecting associations between rare variants and complex traits. SKAT provides analytic power and sample-size calculations to guide the design of new sequence association studies. The method is implemented in software and has been applied to real data, including the Dallas Heart Study, to test associations between rare variants and serum triglyceride levels. SKAT is the most powerful test for dichotomous traits and performs well for continuous traits compared to burden tests. The computation time for SKAT is efficient, making it suitable for genome-wide studies. SKAT is a flexible and powerful method for detecting associations between rare and common genetic variants and complex traits in sequencing-based association studies.