Breeding for productivity and welfare of laying hens

Over the past five decades the art of animal breeding has rapidly advanced into an exacting science including such advanced tools as Best Linear Unbiased Prediction (BLUP) and Restricted Maximum Likelihood (REML) estimation of variance components. Unfortunately, in most selection programmes only traits directly related to productivity are considered and, as a result, those breeding programmes ignore traits that impact on animal welfare. The consequences of ignoring animal welfare in breeding programmes are threefold: (i) if higher-producing animals tend to be more competitive, then the effect of selection is to increase competition, which worsens the animals' welfare; (ii) increased competition has the effect of lowering productivity of other animals that are in direct contention, thus resulting in reduced (or negative) gains for productivity; and (iii) genotype-genotype interaction (competition) invalidates the traditional BLUP animal model and negates many advantages of this technology and could in fact make it a liability. This chapter reviews recent advances that allow selection on production traits but take into account competitive interactions. Those advances fall into two categories: (i) group selection and (ii) advanced mixed models with a second random effect for indirect genetic effects (competition). Both methods have been verified in poultry breeding applications and have resulted in dramatic improvements in livability, productivity and welfare.