Improvements in genetic evaluation for swine

In the first study, reproductive field data from the National Swine Registry was used to estimate breed-specific adjustment factors for number born alive and 21-day litter weight in Duroc, Hampshire, Landrace, and Yorkshire swine. An animal model approach which allowed for simultaneous inclusion of genetic merit with parity, age within parity, number after transfer, and weaning age was utilized. Adjustment factors differed from those previously estimated for all breed and trait combinations and revealed a significant influence of age within parity on parity 1 and 2 litter records.; In the second study, field data from the National Swine Registry was used to estimate additive and dominance variances for Duroc, Hampshire, and Yorkshire swine. Utilizing the parental sub-class dominance relationship matrix and Method R, estimates of dominance variance, expressed as a percentage of phenotypic variance, were found to be approximately 5% for number born alive and 21-day litter weight, approximately 10% for days to 104.5 kg., and 4% for backfat at 104.5 kg. In addition, the regression of phenotypic performance on inbreeding percentage was found to {dollar}-{dollar}.20 for number born alive, {dollar}-{dollar}.50 kg for litter weight, 2.0 for days to 104.5 kg, and 0.0 for backfat per 10% inbreeding.; The final study examined relative changes in prediction of additive breeding values due to inclusion of dominance and/or a regression on inbreeding percentage. On average, the inclusion of either or both of these effects resulted in little change. However, certain sub-sets of the population did reveal differences. The inclusion of a covariate for inbreeding influenced change in inbred animals or animals with inbred progeny for the traits of number born alive, litter weight, and days to 104.5 kg. The inclusion of dominance genetic effects appeared to cause smaller amounts of change although some differences were observed for 21-day litter weight.