scholarly journals Detecting Bias in Large-Scale Comparative Analyses: Methods for Expanding the Scope of Hypothesis-Testing with HormoneBase

2018 ◽  
Vol 58 (4) ◽  
pp. 720-728 ◽  
Author(s):  
Michele A Johnson ◽  
Clinton D Francis ◽  
Eliot T Miller ◽  
Cynthia J Downs ◽  
Maren N Vitousek
Keyword(s):  
Hypothesis Testing ◽  
Large Scale ◽  
Comparative Analyses

scholarly journals 297 GWAS for complex models accounting for populations structure with GBLUP and ssGBLUP

2020 ◽  
Vol 98 (Supplement_4) ◽  
pp. 32-32
Author(s):  
Juan P Steibel ◽  
Ignacio Aguilar
Keyword(s):  
Hypothesis Testing ◽  
Large Scale ◽  
Mixed Model ◽  
Prediction Models ◽  
Association Studies ◽  
Least Square ◽  
Type I ◽  
Phenotypic Variance ◽  
Genome Wide Association Studies ◽  
Formal Hypothesis Testing

Abstract Genomic Best Linear Unbiased Prediction (GBLUP) is the method of choice for incorporating genomic information into the genetic evaluation of livestock species. Furthermore, single step GBLUP (ssGBLUP) is adopted by many breeders’ associations and private entities managing large scale breeding programs. While prediction of breeding values remains the primary use of genomic markers in animal breeding, a secondary interest focuses on performing genome-wide association studies (GWAS). The goal of GWAS is to uncover genomic regions that harbor variants that explain a large proportion of the phenotypic variance, and thus become candidates for discovering and studying causative variants. Several methods have been proposed and successfully applied for embedding GWAS into genomic prediction models. Most methods commonly avoid formal hypothesis testing and resort to estimation of SNP effects, relying on visual inspection of graphical outputs to determine candidate regions. However, with the advent of high throughput phenomics and transcriptomics, a more formal testing approach with automatic discovery thresholds is more appealing. In this work we present the methodological details of a method for performing formal hypothesis testing for GWAS in GBLUP models. First, we present the method and its equivalencies and differences with other GWAS methods. Moreover, we demonstrate through simulation analyses that the proposed method controls type I error rate at the nominal level. Second, we demonstrate two possible computational implementations based on mixed model equations for ssGBLUP and based on the generalized least square equations (GLS). We show that ssGBLUP can deal with datasets with extremely large number of animals and markers and with multiple traits. GLS implementations are well suited for dealing with smaller number of animals with tens of thousands of phenotypes. Third, we show several useful extensions, such as: testing multiple markers at once, testing pleiotropic effects and testing association of social genetic effects.

Large-Scale Hypothesis Testing

2013 ◽  
pp. 15-29 ◽  
Author(s):  
Bradley Efron
Keyword(s):  
Hypothesis Testing ◽  
Large Scale

scholarly journals Illuminating Endocrine Evolution: The Power and Potential of Large-Scale Comparative Analyses

2018 ◽  
Vol 58 (4) ◽  
pp. 712-719 ◽  
Author(s):  
Maren N Vitousek ◽  
Michele A Johnson ◽  
Jerry F Husak
Keyword(s):  
Large Scale ◽  
Comparative Analyses

Hypothesis Testing in Large-scale Functional Linear Regression

2021 ◽  
Author(s):  
Kaijie Xue ◽  
Fang Yao
Keyword(s):  
Linear Regression ◽  
Hypothesis Testing ◽  
Large Scale ◽  
Functional Linear Regression

scholarly journals Large-scale multiple hypothesis testing with the normal-beta prime prior

Statistics ◽  
2019 ◽  
Vol 53 (6) ◽  
pp. 1210-1233 ◽  
Author(s):  
Ray Bai ◽  
Malay Ghosh
Keyword(s):  
Hypothesis Testing ◽  
Large Scale ◽  
Multiple Hypothesis Testing ◽  
Multiple Hypothesis
Sign in / Sign up

Export Citation Format

Share Document