Multi-breed multi-trait single-step genomic predictions for Holstein and Jersey including crossbred animals

Authors

  • Renzo Bonifazi Animal Breeding and Genomics, Wageningen University & Research, P.O. Box 338, 6700 AH Wageningen, The Netherlands https://orcid.org/0000-0002-1794-4708
  • Stella Aivazidou Animal Breeding and Genomics, Wageningen University & Research, P.O. Box 338, 6700 AH Wageningen, The Netherlands
  • Jan ten Napel Animal Breeding and Genomics, Wageningen University & Research, P.O. Box 338, 6700 AH Wageningen, The Netherlands
  • Matias Schrauf Animal Breeding and Genomics, Wageningen University & Research, P.O. Box 338, 6700 AH Wageningen, The Netherlands
  • Gerben de Jong AEU CRV u.a., P.O. Box 454, 6800 AL Arnhem, the Netherlands
  • Lisette Wiesenekker AEU CRV u.a., P.O. Box 454, 6800 AL Arnhem, the Netherlands
  • Jeremie Vandenplas Animal Breeding and Genomics, Wageningen University & Research, P.O. Box 338, 6700 AH Wageningen, The Netherlands

Abstract

Crossbreeding exploits heterozygosity and is increasingly adopted in dairy cattle. However, genomic selection for crossbred animals is challenging due to difficulties in establishing suitable multi-breed reference populations and modelling missing pedigree information. This study aimed to investigate the benefits of multi-breed multi-trait single-step genomic evaluations that jointly analyse New Zealand data from two purebred populations (Holstein and Jersey) and a derived crossbred population (XBD). We also investigated the impact of modelling missing pedigree information using genetic groups (GG) or metafounders (MF). Pedigree (1.1M), genotypes (127K), and individual phenotypes for calving season days (deviation between planned and actual calving date, CSD; ~370K records) and 305-days milk yield (MY; ~538K records) were available for purebred and crossbred animals. Six scenarios were implemented: A) a single-step evaluation per breed, each using phenotypes of all breeds treated as a single trait, but only genotypes of the respective breed, and 255 GG; B) a joint evaluation using the genotypes of all breeds, with phenotypes and GG as in A; C) as B but grouping all GG into only 4 GG; D) as B but replacing all GG by MF; E) as B but replacing all GG by only 4 MF; F) as B but with phenotypes from different breeds treated as separate correlated traits. CSD and MY were jointly analysed in a multi-trait model in all scenarios. Validation statistics were computed for both purebred and XBD genotyped cows and bulls born in recent years. Scenarios using all purebred and XBD genotypes had higher accuracies than the scenario analysing each breed separately. Using all genotypes and modelling traits across breeds as different traits showed the highest accuracy among all scenarios for MY but the lowest for CSD. Reducing the number of GG gave similar results to using all GG. Moving from GG to MF had limited benefits. Overall, results showed that combining Holstein, Jersey, and the derived XBD data into multi-breed single-step evaluations can enhance the accuracy of genomic predictions for both purebred and crossbred animals.

Downloads

Published

2024-09-04

Issue

No. 60 (2024): Proceedings of the 2024 Interbull Meeting

Section

Articles

License

Authors who publish with this journal agree to the following terms:

    1. Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.

 

    1. Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.

 

    1. Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).