The paper introduces BOLT-LMM, an efficient Bayesian mixed model association method that significantly increases the power of genetic association studies, especially in large cohorts. Traditional linear mixed models are powerful for identifying genetic associations and controlling for confounding factors, but they are computationally expensive and assume an infinitesimal genetic architecture. BOLT-LMM addresses these limitations by using a Bayesian mixture prior on marker effect sizes, allowing for more realistic modeling of non-infinitesimal genetic architectures. The method requires only O(MN)-time iterations and scales well with cohort size, making it suitable for large-scale genome-wide association studies (GWAS). The authors demonstrate the effectiveness of BOLT-LMM through simulations and a real-world application to nine quantitative traits in the Women's Genome Health Study (WGHS), showing significant increases in association power. The method is particularly beneficial for traits with a few thousand causal loci and can increase effective sample size by up to 10%. BOLT-LMM is recommended for analyzing randomly ascertained quantitative traits and large, non-ascertained population cohorts, while caution is advised when analyzing ascertained case-control data sets for rare diseases.The paper introduces BOLT-LMM, an efficient Bayesian mixed model association method that significantly increases the power of genetic association studies, especially in large cohorts. Traditional linear mixed models are powerful for identifying genetic associations and controlling for confounding factors, but they are computationally expensive and assume an infinitesimal genetic architecture. BOLT-LMM addresses these limitations by using a Bayesian mixture prior on marker effect sizes, allowing for more realistic modeling of non-infinitesimal genetic architectures. The method requires only O(MN)-time iterations and scales well with cohort size, making it suitable for large-scale genome-wide association studies (GWAS). The authors demonstrate the effectiveness of BOLT-LMM through simulations and a real-world application to nine quantitative traits in the Women's Genome Health Study (WGHS), showing significant increases in association power. The method is particularly beneficial for traits with a few thousand causal loci and can increase effective sample size by up to 10%. BOLT-LMM is recommended for analyzing randomly ascertained quantitative traits and large, non-ascertained population cohorts, while caution is advised when analyzing ascertained case-control data sets for rare diseases.