A Comprehensive Comparison of Haplotype-Based Single-Step Genomic Predictions in Livestock Populations With Different Genetic Diversity Levels: A Simulation Study

Abstract

The level of genetic diversity in a population is inversely proportional to the linkage disequilibrium (LD) between individual single nucleotide polymorphisms (SNPs) and quantitative trait loci (QTLs), leading to lower predictive ability of genomic breeding values (GEBVs) in high genetically diverse populations. Haplotype-based predictions could outperform individual SNP predictions by better capturing the LD between SNP and QTL. Therefore, we aimed to evaluate the accuracy and bias of individual-SNP- and haplotype-based genomic predictions under the single-step-genomic best linear unbiased prediction (ssGBLUP) approach in genetically diverse populations. We simulated purebred and composite sheep populations using literature parameters for moderate and low heritability traits. The haplotypes were created based on LD thresholds of 0.1, 0.3, and 0.6. Pseudo-SNPs from unique haplotype alleles were used to create the genomic relationship matrix (G) in the ssGBLUP analyses. Alternative scenarios were compared in which the pseudo-SNPs were combined with non-LD clustered SNPs, only pseudo-SNPs, or haplotypes fitted in a second G (two relationship matrices). The GEBV accuracies for the moderate heritability-trait scenarios fitting individual SNPs ranged from 0.41 to 0.55 and with haplotypes from 0.17 to 0.54 in the most (Ne ≅ 450) and less (Ne < 200) genetically diverse populations, respectively, and the bias fitting individual SNPs or haplotypes ranged between −0.14 and −0.08 and from −0.62 to −0.08, respectively. For the low heritability-trait scenarios, the GEBV accuracies fitting individual SNPs ranged from 0.24 to 0.32, and for fitting haplotypes, it ranged from 0.11 to 0.32 in the more (Ne ≅ 250) and less (Ne ≅ 100) genetically diverse populations, respectively, and the bias ranged between −0.36 and −0.32 and from −0.78 to −0.33 fitting individual SNPs or haplotypes, respectively. The lowest accuracies and largest biases were observed fitting only pseudo-SNPs from blocks constructed with an LD threshold of 0.3 (p < 0.05), whereas the best results were obtained using only SNPs or the combination of independent SNPs and pseudo-SNPs in one or two G matrices, in both heritability levels and all populations regardless of the level of genetic diversity. In summary, haplotype-based models did not improve the performance of genomic predictions in genetically diverse populations.

Expanding the application of haplotype-based genomic predictions to the wild: A case of antibody response against Teladorsagia circumcincta in Soay sheep

“…Regardless of the applied methods, considerable improvement in the accuracy of GEBVs was observed for GPs of IgA based on the combination of SNPs and non-LD clustered SNPs compared to GBLUP using individual SNPs, with the highest accuracy obtained by A COM0.15 . Our results are in disagreement with the previous study in which the combinations of SNPs and non-clustered pseudo-SNPs were used for GP for the rst time and showed no improvement in accuracy [23].…”

“…Haplotypes are more informative than SNPs in describing recent identical-by-descent (IBD) relationships, and they may also be more effective in capturing linkage disequilibrium (LD) with multiallelic QTL than individual SNPs, which are often biallelic [19]. In practice, the performance of GP based on haplotypes varied across traits and species, ranging from negligible to substantial increases in accuracy compared to SNP-based models [19,[21][22][23][24]. Three different methods were applied to de ne haploblocks, including (i) a xed number of SNPs per haplotype block [25], (ii) xed block length [26], and (iii) LD blocks [27].…”

“…Due to the presence of relatively strong LD among particular markers, the number of variants per haploblock is reduced considerably compared to when haploblocks are de ned by a xed number of close SNPs [27]. Haplotypic information could be then integrated as pseudo-SNP into BLUP [21,22] or Bayesian [26] GP models, or based on a recent method applied by Araujo et al [23], the pseudo-SNPs can contribute to a genomic relationship matrix (GRM) construction in combination with non-LD clustered SNPs, i.e., those located out of haploblocks.…”

Impact of linkage disequilibrium heterogeneity along the genome on genomic prediction and heritability estimation

“…In general, LD information between markers is used to pre-select markers or construct LD-based haplotypes, but these processes have little impact on the accuracy of genomic prediction [ 6 – 8 ]. However, another factor that seems to have a greater impact on genomic prediction is the uneven distribution of LD along the genome, i.e.…”

Editorial: Beef on Dairy: The Use of a Simple Tool to Improve Both Cattle Production Systems

“…This special Research Topic includes six papers presenting novel and exciting breeding strategies and methods such as genomic prediction of breeding values, genome-wide association studies, beef-cross-dairy performance, and reproductive technologies. Fitting haplotypes instead of single nucleotide polymorphism (SNP) markers in genomic prediction models was investigated as an approach to improve the accuracy of genomic breeding values ( Feitosa et al, 2020 ; Teissier et al, 2020 ; Araujo et al, 2021 ). In this context, Li et al evaluated the predictive performance of carcass traits in Chinese Simmental cattle using haplotypes in comparison to the traditional genomic prediction approaches.…”