Effect of service sire and validity of repeatability model in litter size and farrowing interval of Finnish Landrace and Large White populations

Abstract

Data from the Finnish litter-recording scheme were used to find out the effect of service sire on litter size, and to determine whether the assumptions of repeatability model are filled or not in the breeding value estimation of litter size and farrowing interval. Data contained litter records of 6514 Finnish Large White and 9154 Landrace pigs born between 1985 and 2000. Pedigree information was traced back to 1983. (Co)variance components were estimated by the restricted maximum likelihood method using three different statistical models. Variances due to direct animal effects and the service sire, and the covariance between them, were analysed initially using paternal effect model for the first parity litter size records. The same approach was subsequently extended to multiple records per sow by including the permanent environment for dams also. Genetic correlations among different parities were estimated using a multitrait animal model without service sire effects. The effects of model modifications on ranking of selection candidates were studied by comparing pedigree indices of young boars born in 1998 or later. Estimated variance components for both traits were similar across breeds. Heritability estimates ranged between 0.11 and 0.15 for total number of piglets born per litter and between 0.04 and 0.16 for farrowing interval, with estimates being considerably higher within the first two parities than later parities. The proportions of variance due to service sire effect on litter size were extremely low, and the genetic correlation between direct and service sire effects varied around zero. In spite of close correlations of litter sizes across different parities, it appeared that the genetic correlations between the first and later were lower than the correlations between later parities. All genetic correlations for farrowing interval among different parities were lower than one. In conclusion, the first and later parities of litter size, and all the parities of farrowing interval should be treated as separate traits in breeding value estimation. Although the effect of service sire is low, it should still be considered in the statistical model as a random environmental effect to get efficient tool for phenotypic selection of AI boars due to poor piglet production.