Genetic analyses of piglet survival and individual birth weight on first generation data of a selection experiment for piglet survival under outdoor conditions

Genetic parameters of piglet survival traits and birth weight were estimated on the first generation data of a selection experiment aimed at improving piglet survival using a multiple trait linear and threshold model. Data on 5293 piglets for survival at birth, at day one after birth and during the entire nursing period, as well as individual birth weight and litter size, were recorded in an outdoor production system. Genetic effects of piglet survival traits and birth weight were estimated based on threshold and Gaussian models, respectively, using a Bayesian approach. The statistical model included as fixed effects selection group, parity, gender, fostering, gestation length and month of farrowing and, alternatively, an adjustment for litter size. Direct genetic effects (i.e. the piglet's genetic potential) for piglet survival and birth weight were estimated separately, whereas maternal genetic and environmental effects could only be estimated for the given data structure in a combined litter effect. Posterior means of heritabilities for direct genetic effects of survival at birth, at first day after birth and the entire nursing period, as well as birth weight, were 0.08, 0.07, 0.08 and 0.20, respectively. Genetic correlations among survival traits were in the range of 0.29 to 0.40 and indicate that these traits were mainly attributable to different genetic effects. Genetic correlations between direct effects of survival traits and birth weight ranged between 0.18 and 0.23 and were reduced when weights of stillborn piglets were omitted in the analysis or the traits were adjusted for litter size. The magnitudes of direct genetic effects of survival traits are substantially higher than estimates in the literature, which may indicate that these traits have a higher genetic influence under outdoor conditions. The use of birth weight in the multiple trait estimation provided important information for the estimation of survival traits due to its favourable genetic correlations with survival, its high heritability and its high information content as a continuously measured trait