Marker effect p-values for single-step GWAS with the algorithm for proven and young in large genotyped populations

Area of Science:

• Animal Breeding and Genetics
• Bioinformatics
• Statistical Genetics

Background:

• Single-nucleotide polymorphism (SNP) effects are crucial for genomic estimated breeding values (GEBV) and genome-wide association studies (GWAS).
• Traditional methods for calculating SNP p-values in GWAS involve computationally intensive matrix inversions, limiting their application in large populations.
• A novel approximation method is proposed to address these computational challenges.

Purpose of the Study:

• To develop and validate an efficient method for approximating SNP p-values in single-step genomic best linear unbiased prediction (ssGBLUP) based GWAS.
• To reduce the computational burden associated with calculating SNP p-values in large genotyped populations.
• To enhance the power of GWAS by enabling analysis of larger datasets.

Main Methods:

• Utilized a sparse approximation of the inverse genomic relationship matrix (G) with the Algorithm for Proven and Young (APY).
• Employed an approximation of the prediction error variance for SNP effects, avoiding direct inversion of the left-hand side (LHS) of mixed model equations.
• Validated the method on a 50K genotyped animal population and applied it to a 450K genotyped animal population.

Main Results:

• The approximation method identified the same significant genomic regions on chromosomes 7 and 20 as benchmark methods in the 50K population.
• Computational time was reduced by 38-fold, and memory requirements decreased tenfold compared to exact inversion.
• Analysis of the 450K population revealed two novel significant regions on chromosomes 6 and 14, demonstrating increased GWAS detection power.

Conclusions:

• The proposed method makes obtaining SNP p-values for ssGWAS computationally feasible in large genotyped populations.
• Computational cost is no longer a bottleneck for conducting GWAS in extensive populations with numerous genotyped animals.
• This advancement facilitates more comprehensive genetic analyses and marker discovery.