Single- and multiple-breed genomic evaluations for conformation traits in Canadian Alpine and Saanen dairy goats
Erin Massender,Luiz F Brito,Laurence Maignel,Hinayah R Oliveira,Mohsen Jafarikia,Christine F Baes,Brian Sullivan,Flavio S Schenkel
DOI: https://doi.org/10.3168/jds.2021-21713
Abstract:Conformation traits are functional traits known to affect longevity, production efficiency, and profitability of dairy goats. However, genetic progress for these traits is expected to be slower than for milk production traits due to the limited number of herds participating in type classification programs, and often lower heritability estimates. Genomic selection substantially accelerates the rate of genetic progress in many species and industries, especially for lowly heritable, difficult, or expensive to measure traits. Therefore, the main objectives of this study were (1) to evaluate the potential benefits of the implementation of single-step genomic evaluations for conformation traits in Canadian Alpine and Saanen dairy goats, and (2) to investigate the effect of the use of single- and multiple-breed training populations. The phenotypes used in this study were linear conformation scores, on a 1-to-9 scale, for 8 traits (i.e., body capacity, dairy character, fore udder, feet and legs, general appearance, rear udder, medial suspensory ligament, and teats) of 5,158 Alpine and 2,342 Saanen does. Genotypes were available for 833 Alpine and 874 Saanen animals. Averaged across all traits, the use of multiple-breed analyses increased validation accuracy for Saanen, and reduced bias of genomically enhanced breeding values (GEBV) for both Alpine and Saanen compared with single-breed analyses. Little benefit was observed from the use of GEBV relative to pedigree-based EBV in terms of validation accuracy and bias, possibly due to limitations in the validation design, but substantial gains of 0.14 to 0.21 (32-50%) were observed in the theoretical accuracy of validation animals when averaged across traits for single- and multiple-breed analyses. Across the whole genotyped population, average gains in theoretical accuracy for GEBV compared with EBV across all traits ranged from 0.15 to 0.17 (32-37%) for Alpine and 0.17 to 0.19 (40-41%) for Saanen, depending on the model used. The largest gains were observed for does without classification records (0.19-0.22 or 50-55%) and bucks without daughter classification records (0.20-0.27 or 57-82%), which have the least information contributing to their traditional EBV. The use of multiple-breed rather than single-breed models was most beneficial for the Saanen breed, which had fewer phenotypic records available for the analyses. These results suggest that the implementation of genomic selection could increase the accuracy of breeding values for conformation traits in Canadian dairy goats.