Resampling-based tests for Lasso in genome-wide association studies

BMC Genetics

Methodology Article

Matt McGue¹  Snigdhansu Chatterjee²  Saonli Basu³  Jaron Arbet³ 

[1] Department of Psychology, University of Minnesota, 55455, Minneapolis, USA;Department of Statistics, University of Minnesota, 55455, Minneapolis, USA;Division of Biostatistics, School of Public Health, University of Minnesota, 55455, Minneapolis, USA;

关键词: Lasso;    GWAS;    Resampling;    Permutation;    Bootstrap;    Testing;   

DOI  :  10.1186/s12863-017-0533-3

 received in 2016-12-05, accepted in 2017-06-30,  发布年份 2017

来源: Springer

【 摘 要 】

BackgroundGenome-wide association studies involve detecting association between millions of genetic variants and a trait, which typically use univariate regression to test association between each single variant and the phenotype. Alternatively, Lasso penalized regression allows one to jointly model the relationship between all genetic variants and the phenotype. However, it is unclear how to best conduct inference on the individual Lasso coefficients, especially in high-dimensional settings.MethodsWe consider six methods for testing the Lasso coefficients: two permutation (Lasso-Ayers, Lasso-PL) and one analytic approach (Lasso-AL) to select the penalty parameter for type-1-error control, residual bootstrap (Lasso-RB), modified residual bootstrap (Lasso-MRB), and a permutation test (Lasso-PT). Methods are compared via simulations and application to the Minnesota Center for Twins and Family Study.ResultsWe show that for finite sample sizes with increasing number of null predictors, Lasso-RB, Lasso-MRB, and Lasso-PT fail to be viable methods of inference. However, Lasso-PL and Lasso-AL remain fast and powerful tools for conducting inference with the Lasso, even in high-dimensions.ConclusionOur results suggest that the proposed permutation selection procedure (Lasso-PL) and the analytic selection method (Lasso-AL) are fast and powerful alternatives to the standard univariate analysis in genome-wide association studies.

【 授权许可】

CC BY   
© The Author(s) 2017

【 图 表 】

12864_2015_2055_Article_IEq29.gif

12864_2015_2192_Article_IEq22.gif

12864_2017_3771_Article_IEq13.gif

12903_2017_385_Article_IEq5.gif

12864_2017_3610_Article_IEq1.gif

12888_2016_877_Article_IEq15.gif

12864_2017_3733_Article_IEq22.gif

12864_2017_4004_Article_IEq14.gif

12898_2017_155_Article_IEq33.gif

12864_2015_2300_Article_IEq26.gif

12864_2015_2300_Article_IEq24.gif

12864_2017_3733_Article_IEq17.gif

12864_2017_3487_Article_IEq15.gif

12870_2017_1059_Article_IEq7.gif

12864_2017_3547_Article_IEq1.gif

12864_2015_2192_Article_IEq11.gif

12888_2016_877_Article_IEq6.gif

12864_2017_3669_Article_IEq5.gif

12888_2016_877_Article_IEq4.gif

12864_2016_2789_Article_IEq16.gif

12888_2016_877_Article_IEq3.gif

12864_2017_4132_Article_IEq3.gif

12864_2017_4186_Article_IEq28.gif

12864_2017_4132_Article_IEq2.gif

12864_2017_4132_Article_IEq1.gif

12864_2017_4186_Article_IEq26.gif

12888_2016_811_Article_IEq1.gif

12864_2017_4186_Article_IEq25.gif

12864_2017_4186_Article_IEq23.gif

12864_2017_4030_Article_IEq18.gif

12864_2016_3477_Article_IEq13.gif

12864_2017_4004_Article_IEq5.gif

12864_2017_4186_Article_IEq21.gif

12864_2017_4186_Article_IEq34.gif

12864_2016_2682_Article_IEq54.gif

12864_2017_4004_Article_IEq1.gif

12864_2016_2796_Article_IEq12.gif

12864_2015_2055_Article_IEq94.gif

12864_2016_3098_Article_IEq60.gif

12864_2016_2821_Article_IEq44.gif

12864_2017_4274_Article_IEq10.gif

12888_2017_1504_Article_IEq3.gif

12888_2017_1504_Article_IEq2.gif

12864_2016_3477_Article_IEq5.gif

12864_2016_2821_Article_IEq37.gif

12864_2017_3500_Article_IEq22.gif

12864_2017_3910_Article_IEq1.gif

12864_2017_4000_Article_IEq1.gif

12864_2015_2055_Article_IEq79.gif

12864_2017_3744_Article_IEq1.gif

12864_2016_3440_Article_IEq34.gif

12864_2016_3440_Article_IEq32.gif

12864_2017_4025_Article_IEq10.gif

12864_2017_4025_Article_IEq9.gif

12864_2017_4025_Article_IEq8.gif

12864_2017_4025_Article_IEq7.gif

12864_2017_4025_Article_IEq6.gif

12864_2017_4025_Article_IEq5.gif

12864_2017_4025_Article_IEq4.gif

12864_2016_2443_Article_IEq15.gif

12864_2015_2055_Article_IEq66.gif

12864_2017_4025_Article_IEq2.gif

12864_2015_2118_Article_IEq5.gif

12864_2017_3785_Article_IEq2.gif

12864_2017_3604_Article_IEq5.gif

12864_2016_3263_Article_IEq15.gif

12864_2017_3604_Article_IEq3.gif

12864_2016_2789_Article_IEq51.gif

12864_2016_2789_Article_IEq49.gif

12864_2016_2821_Article_IEq13.gif

12888_2017_1557_Article_IEq9.gif

12864_2017_3783_Article_IEq1.gif

12864_2017_4309_Article_IEq7.gif

12902_2017_161_Article_IEq1.gif

12864_2017_3655_Article_IEq8.gif

12864_2016_2789_Article_IEq44.gif

12864_2015_2001_Article_IEq2.gif

12864_2017_3733_Article_IEq63.gif

12864_2017_3733_Article_IEq62.gif

12864_2016_3098_Article_IEq29.gif

12864_2016_2463_Article_IEq1.gif

12864_2017_3733_Article_IEq57.gif

12864_2016_3098_Article_IEq87.gif

12864_2017_3733_Article_IEq56.gif

12864_2017_3733_Article_IEq55.gif

12864_2017_3781_Article_IEq9.gif

12864_2017_3521_Article_IEq4.gif

12864_2017_4020_Article_IEq15.gif

12864_2017_3521_Article_IEq3.gif

12864_2017_4132_Article_IEq36.gif