Estimation of individual animal SNP-BLUP reliability using full Monte Carlo sampling

H Ben Zaabza,E.A Mäntysaari,I Strandén

doi:10.3168/jdsc.2020-0058

H Ben Zaabza, E.A Mäntysaari + Show 1 more

Open Access

https://doi.org/10.3168/jdsc.2020-0058

Copy DOI

Journal: JDS Communications	Publication Date: Mar 19, 2021
Citations: 7	License type: cc-by

Affiliation: Natural Resources Institute Finland

Abstract

Calculation of individual animal reliability of estimated genomic breeding value by SNP-BLUP requires inversion of the mixed model equations (MME). When the SNP-BLUP model includes a residual polygenic (RPG) effect, the size of the MME will be at least the number of genotyped animals (n) plus the number of SNP markers (m). Inversion of the MME in SNP-BLUP involves computations proportional to the cube of the MME size; that is, (n + m)3, which can present a considerable computational burden. We introduce a full Monte Carlo (MC) sampling-based method for approximating reliability in the SNP-BLUP model and compare its performance to the genomic BLUP (GBLUP) model. The performance of the full MC approach was evaluated using 2 data sets, including 19,757 and 222,619 genotyped animals selected from populations with 231,186 and 13.35 million pedigree animals, respectively. Genotypes were available in the data sets for 11,729 and 50,240 SNP markers. An advantage of the full MC approximation method was its low computational demand. A drawback was its tendency to overestimate reliability for animals with low reliability, especially when the weight of the RPG effect was high. The overestimation can be lessened by increasing the number of MC samples.

Highlights

We introduce a full Monte Carlo (MC) sampling-based method for approximating reliability in the SNP-BLUP model and compare its performance to the genomic BLUP (GBLUP) model
The relative difference was even larger when the residual polygenic (RPG) weight was 80% (188 × 10−5 vs. 8,280 × 10−5). These results suggest that the full-MC-SNPBLUP model approximation is less sensitive, in terms of mean-squared error (MSE), to changes from low to high RPG weight than the MC-SNP-BLUP approach at the same mixed model equations (MME) sizes
In the scenario with an 80% RPG weight, the slope was 1.70 with MC-SNP-BLUP (10,000 MC samples) but reduced to 1.08 with full-MC-SNPBLUP (20,000 MC samples). This was partly a consequence of comparing equal-size MME results where the number of MC samples used for approximating the RPG effect was larger in fullMC-SNP-BLUP than in MC-SNP-BLUP

Summary

Introduction

When the SNP-BLUP model includes a residual polygenic (RPG) effect, the size of the MME will be at least the number of genotyped animals (n) plus the number of SNP markers (m). The full-MC-SNP-BLUP model was used to calculate approximated reliabilities under 10 different MC sample sizes (10,000, 15,000, 20,000, 30,000, 40,000, 50,000, 60,000, 70,000, and 90,000).

Results

Conclusion