Differential handling of missing parents in genetic evaluation of dairy cattle using single-step test-day SNP-BLUP model

Abstract

Single-step genomic models use all available information on animals’ phenotype, genotype and pedigree. Nowadays, many countries aim towards implementing single-step models and replacing the existing conventional models for routine evaluation. Even in the area of genomic evaluation, the pedigree data has still a significant impact on estimated genomic breeding values, and therefore it is very important to obtain the most informative structure of the pedigree. The crucial aspect of the pedigree editing is handling missing parents information. Missing data can arise either due to truly missing parentage information, or due to the fact that not all generations are utilized. We focused on three scenarios for handling missing parents: 1) raw pedigree, where missing parents IDs were set to missing; 2) genetic groups, where missing parents in the raw pedigree were replaced by genetic groups based on year of birth, country of origin, and sex; 3) metafounders, which are created based on genetic groups and genomic information. The genomic breeding values for fat yield were estimated using the single-step test-day SNP-BLUP model implemented by the MiXBLUP software. The analysed data corresponds to the population of Polish Holstein Friesian cattle used for routine genetic evaluation. We compared the results of the validation obtained by the three pedigree handling approaches and observed that the best results of validation were achieved by the scenario with metafounders (3), followed by scenario fitting pedigree with genetic groups (2), and finally by the raw pedigree (1). The metafounders scenario uses most of the information including genotype data, therefore, it provides the best classification of unknown animals into groups, which improves validation results.