The relationship between reproductive performance of crossbred sows and the EBVs of their purebred dams

In vertically integrated pig production schemes the number of crossbred sows at the commercial level is much higher than the number of purebred sows at the nucleus and multiplier levels. This indicates the relative importance of enhancing the performance of the crossbred offspring rather than that of the purebreds themselves and reinforces the idea of setting the breeding goal towards enhancing the performance at the commercial level (Brandt and Täubert, 1998). The estimated breeding values (EBVs) obtained for a certain trait in purebred dams at the nucleus and multiplier levels using only purebred information, reflect the additive genetic effects of the purebred animals and can be used to predict the genetic merit of their purebred daughters. However, do EBVs from purebred dams predict the performance of their crossbred daughters in the commercial environment? Genetic correlations between purebred and crossbred sows for reproductive traits have been estimated previously and estimates ranged from 0.21 to 0.99 in the studies by Täubert et al. (1998), Boesch et al. (1998) and Nakavisut et al. (2005). Such a spread of estimates makes it difficult to extrapolate results to a specific case. A full pedigree structure at the commercial level is required to estimate these genetic correlations. If full pedigree information is not available, performance of crossbred offspring may be regressed on the EBV of a parent. This approach was used by Hall et al. (2002) in sheep data. The aim of our work was to study the relationship between the reproductive performance of commercial crossbred F1 sows and the EBVs of their purebred dams based on purebred data only.