# Weighted GBLUP in Simulated Beef Cattle Populations: Impact of Reference Population, Marker Density, and Heritability

## TL;DR

This study shows that using weighted genomic BLUP improves the accuracy of predicting genetic traits in beef cattle, especially when marker density is high.

## Contribution

The study introduces the effectiveness of weighted GBLUP in enhancing genomic prediction accuracy in beef cattle under varying genetic conditions.

## Key findings

- wGBLUP improves prediction accuracy for low-heritability traits in small, high-density marker populations.
- Increasing marker density and reference population size enhances genomic prediction accuracy.
- Integrating pedigree, genomic, and weighted SNP data reduces bias in genetic evaluation.

## Abstract

Genomic selection (GS) enhances breeding efficiency by integrating genomic data with pedigree information and phenotypes. Its effectiveness varies among livestock, with beef cattle facing challenges due to breed diversity. Therefore, this study aims to evaluate the impact of different levels of heritability, marker densities, and selection designs on the accuracy of genomic prediction in multiple beef cattle breeds through simulation studies, comparing the predictive accuracy of different methods such as PBLUP, GBLUP, and wGBLUP in simulated populations, with the goal of improving the accuracy of GP in beef cattle across different genetic backgrounds. Ultimately, we found that the use of the wGBLUP method can significantly enhance the accuracy of GP. These findings are crucial for optimizing GS in beef cattle breeding.

Genomic selection (GS) is a technique that integrates genomic data, pedigree information, and individual phenotypes to enhance genetic improvements of economically important traits in livestock. While it has shown significant effects in dairy cattle, its efficacy in beef cattle is lower due to breed diversity and differences in reproductive structures. Therefore, this study evaluated the impact of heritability levels, marker densities, and assessment methods (such as pedigree-based BLUP, genomic BLUP, and weighted genomic BLUP) on genomic prediction accuracy across multiple beef cattle breeds through simulations. Three beef cattle populations were simulated with heritability levels set at 0.3, 0.5, and 0.7 and marker densities set at 50 k and 770 k. The results showed that the predictive accuracy of PBLUP and GBLUP increased with higher heritability and larger reference populations. Increasing the marker density also improved the accuracy of genomic predictions; even a low marker density (50 k SNP) can significantly enhance the accuracy of genetic evaluation, although the size of the reference population needs to be optimized according to population structure, heritability, and the genetic architecture of the trait. Overall, integrating pedigree, genomic, and weighted SNP information can significantly improve the precision of GEBV prediction and reduce bias. In particular, the wGBLUP method demonstrated an improvement in the prediction accuracy of low-heritability traits in small but high-density marker populations.