Bayesian Analysis of Genetic Associations of Skin Lesions and Behavioural Traits to Identify Genetic Components of Individual Aggressiveness in Pigs
There is increasing interest in genetic selection against behavioural traits that impact negatively on welfare and productivity in commercial livestock production. Post-mixing aggressiveness in pigs shows wide phenotypic variation, affects health, welfare and growth performance and is a routine feature of production. A Bayesian approach was used to estimate the heritability of three traits associated with aggressiveness in pigs during the 24 h post-mixing; duration in reciprocal aggression, and in receipt of, or delivery of non-reciprocal aggression (NRA). For the purposes of genetic selection, recording aggressive behaviour is excessively labour intensive. The genetic correlations were quantified between the behavioural traits and an easily measurable indicator trait; the number of skin lesions following mixing (lesion score, LS). The heritabilities for the three behavioural traits ranged from 0.17 to 0.46 (receipt of NRA and reciprocal aggression respectively). The duration in reciprocal aggression and in delivery of NRA showed a strong genetic correlation (r g = 0.79 with 95% Bayesian credibility interval of 0.62–0.94). The genetic correlation between LS and these two behaviours indicated that selection on breeding values of LS could be used to reduce aggressiveness. The duration in receipt of NRA appeared to be regulated by different genes or genomic effects compared with the other behavioural traits and LS. Although duration in receipt of NRA was not genetically associated with LS, it was lowly but significantly environmentally associated with the residuals of central and caudal LS (r e = 0.28–0.32), indicating that pigs that received NRA also received bites on the central and caudal third of the body. The pen that the animals were mixed into was found to be a very important factor for the analysed traits, in particular those representing behavioural characteristics. Based on the estimated genetic parameters, it is concluded that selection on breeding values for reduced LS (especially central LS) is expected to reduce reciprocal aggression and the delivery of NRA but will not change the receipt of NRA directly.
KeywordsAggression Behaviour Lesion score Genetic correlation Pig
The authors thank the staff of the pig unit for their invaluable assistance. This project was funded through the LINK Sustainable Livestock Production programme by Defra with additional funding from the PIC International Group, JSR Genetics, the Meat and Livestock Commission, the Royal Society for the Prevention of Cruelty to Animals and the Scottish Society for the Prevention of Cruelty to Animals. SAC receives financial support from SEERAD.
- Francis DA, Christison GI, Cymbaluk NF (1996) Uniform or heterogeneous weight groups as factors in mixing weanling pigs. Can J Anim Sci 76:171–176Google Scholar
- Kanis E, van den Belt H, Groen AF, Schakel J, de Greef KH (2004) Breeding for improved welfare in pigs: a conceptual framework and its use in practice. Anim Sci 78:315–329Google Scholar
- Misztal I, Tsuruta S, Strabel T, Auvray B, Druet T, Lee DH (2002) BLUF90 and related programs (BGF90). In: Proceedings of the 7th world congress genetics applied to livestock production, Montpellier, France. CD-ROM. Communication No. 28-07Google Scholar
- Raftery AE, Lewis SM (1992) How many iterations in the Gibbs sampler? In: Bernado JM, Berger JO, Dawid AP, Smith AFM (eds) Bayesian statistics IV. Oxford University Press, Oxford, pp 763–774Google Scholar
- Spoolder HAM, Edwards SA, Corning S (1999) Effects of group size and feeder space allowance on welfare in finishing pigs. Anim Sci 69:481–489Google Scholar
- Thomas DC, Gauderman WJ (1996) Gibbs sampling methods in genetics. In: Gilks WR, Richardson S, Spiegelhalter DJ (eds) Markov chain Monte Carlo in practice. Chapman & Hall, London, pp 419–440Google Scholar