Abstract
For swine breeding programs, testing and selection programs are located in nucleus units that are generally managed differently and with higher health levels than commercial herds where descendants of nucleus animals are expected to perform. This approach assumes that superior animals selected in nucleus herds will have progeny with superior performance at the commercial level. There is clear evidence that this may not be true for all traits of economic importance and thus methods including data collected at the commercial level may increase accuracy of selection at the nucleus level. This study’s goal was to estimate genetic parameters for five reproductive traits between two purebred maternal nucleus populations and their commercial F1 offspring: Total Number Born, Number Born Alive, Number Born Alive > 1kg, Number Weaned, and Litter Weight. Estimates were based on single-step GBLUP in the BLUPF90 programs by utilizing any two combinations of a purebred and the F1 population, and by using all three populations jointly. The genomic relationship matrix between the three populations was generated by using within-population allele frequencies for relationships within a population, and across-population allele frequencies for relationships of the F1 with the purebred animals. The two purebred populations were assumed to be genetically unrelated. The use of two versus three populations did not impact estimates of heritability, additive variance, or genetic correlations. Heritabilities ranged from 0.02 to 0.09 for the F1, from 0.07 to 0.18 for Landrace, and from 0.07 to 0.21 for Yorkshire. Genetic correlations between the same traits in F1 and Landrace ranged from 0.22 to 0.93, and from 0.31 to 0.84 for F1 and Yorkshire. This range of genetic correlations indicates that the use of crossbred information can aid in the selection of purebreds for commercial crossbred performance for some traits. Funded by Genome Canada Genomic Applications Partnerships Program.