Indirect genetic effects for group-housed animals

ABSTRACT Alemu, SW(2015) Indirect Genetic effects for Group-housed Animals. Joint PhD thesis between Aarhus University, Denmark and Wageningen University, the Netherlands. Social interactions among individuals are common both in plants and animals. With social interactions, the trait value of an individual may be influenced by the genes of its interacting partners, a phenomenon known as indirect genetic effects (IGE). An IGE is heritable effect of an individual on trait values of another individual. A large body of literature has shown that social interactions can create addition heritable variation in both plants and animals, for both behavioural and production traits. When IGE are estimated it is usually assumed that an individual interacts equally with all its group mates, irrespective of genetic relatedness. This assumption may not be true in mixed groups of kin and non-kin, where an individual may interact systematically different with kin and non-kin. Current IGE models ignore such systematically different interactions between kin and non-kin. Thus, the main aim of this thesis was to develop and apply statistical methods to estimate IGE when interactions differ between kin and non-kin. Social interactions are important in mink that are kept in groups for the production of fur. Group housing of mink increases aggression behaviours, which is reflected by an increase in the number of bite marks on the pelts, and reduces the welfare of the animals. We estimated the genetic parameter for bite mark traits in group-housed mink, to investigate the prospects for genetic improvement of bite mark traits. We found that there are good prospects to produce mink that have a low level of biting. Finally, we further concluded that genetic parameter estimation for bite mark score should take into account systematic interactions due to sex or kin. In this thesis we also investigated genomic selection for socially affected traits, considering survival time in two lines of brown egg layers showing cannibalistic behaviour. Despite the limited reference population of ~234 progeny tested sires, the accuracy of estimated breeding values (EBV) was ~35% higher for genomic selection compared with the parent average-EBV. We found that the response to genomic selection per year for line B1 was substantially higher than for the traditional breeding scheme, whereas for line BD response was slightly higher than for the traditional breeding scheme. In conclusion, genetic selection with IGE combined with marker information can substantially reduce detrimental social behaviours such as cannibalism in layers and biting in group-housed mink.