Comparison of haplotype method using for genomic prediction versus single SNP genotypes in sheep breeding programs

Linkage disequilibrium (LD) advancement map and the specification of population-level haplotype block structures are parameters that help manage the study of the Genome-wide Association (GWAS), and to comprehend the nature of the non-linear relationship among phenotypes and genotype. Compared with single nucleotide polymorphisms (SNP), genomic prediction fitting haplotype alleles and improve prediction accuracy; but the increase in accuracy belongs to how the Haplotype block is characterized. The Haplotype alleles were defined according to the SNP alleles in not covering blocks 125 Kb, 250 Kb, 500 Kb, and 1 Mb. The Haplotype alleles with frequencies below 1, 2.5, 5, or 10 % are eliminated. Using BayesB for genomic prediction fit covariate for SNP or haplotype. Three traits with three heritability levels of low: 0.10 (milk production (MILK)), medium: 0.30 (carcass weight (CARCASS)) and high: 0.45 (mature body weight (MATURE)) in sheep were assumed to be influencing the trait of interest. Under all situations, the average genomic accuracy increased when heritability values increased. When the haploblock length is properly characterized, covariates that are also suitable in haplotype alleles instead of SNPs will improve genome precision. Fitting a haploblock with a length of 250 kb-1 Mb with a frequency threshold of 1% leads to the highest accuracy of genomic prediction. Increasing the frequency threshold had a more negative impact on both accuracy and bias for the longer haploblocks than shorter haploblocks.

Automated summary

The study found that when the length of haploblocks is properly characterized, covariates suitable for haplotype alleles instead of SNPs will improve the accuracy of the genome. Longer haploblocks are more negatively affected by increasing the frequency threshold, leading to a greater impact on accuracy and bias.