AUTHOR=Ye Haoqiang , Zhang Zipeng , Ren Duanyang , Cai Xiaodian , Zhu Qianghui , Ding Xiangdong , Zhang Hao , Zhang Zhe , Li Jiaqi 

TITLE=Genomic Prediction Using LD-Based Haplotypes in Combined Pig Populations

JOURNAL=Frontiers in Genetics

VOLUME=Volume 13 - 2022

YEAR=2022

URL=https://www.frontiersin.org/journals/genetics/articles/10.3389/fgene.2022.843300

DOI=10.3389/fgene.2022.843300

ISSN=1664-8021

ABSTRACT=The size of reference population is an important factor affecting genomic prediction, so combining different populations in genomic prediction is an attractive way to increase accuracy compared with within-population genomic prediction. However, by simply combining population, the accuracy of genomic prediction has been limited or even slightly decreased, which was probably because the linkage disequilibrium (LD) pattern is different between different populations. Many studies have shown that using whole-genome sequence (WGS) data or haplotype method can improve the accuracy of multi-population genomic prediction by enhancing the ability to capture LD between variants and quantitative trait loci (QTL). In this study, we used imputed whole-genome sequence (WGS) data to construct LD-based haplotypes for genomic prediction in combined populations and the haploblocks were defined based on the different LD thresholds. Our objective was to explore the impact of different SNP densities, variant predictors and reference population sizes on prediction accuracy for reproduction traits in Yorkshire pig populations. Comparing to chip data, genomic best linear unbiased prediction (GBLUP) using WGS data can improve the accuracy of prediction in multi-population but not within-population. The genomic prediction accuracy of haplotype method using 80K chip data in multi-population was not only higher than that within-population (1.2% to 4.3%), but also GBLUP for the multi-population (3.4% to 5.9%). More importantly, We have found that using haplotype method based on WGS data in multi-populations has better genomic prediction performance, and our results had shown that building haploblock in this scenario based on low LD threshold (r2 = 0.2~0.3) produced an optimal set of variables for reproduction traits in Yorkshire pig population. Our results suggested that whether the use of haplotype method based on chip data or GBLUP (individual SNP method) based on WGS data were beneficial for genomic prediction in multi-population, while simultaneously combining haplotype method and WGS data was better genomic evaluation strategy for multi-population.