POWER CALCULATIONS FOR OPTIMISATION OF THE EXPERIMENTAL DESIGN TO DETECT G X E: SALMON EXPERIMENTS IN FUTUREEUAQUA

The objective of the WP1 in FutureEUAqua is to assess how the current breeding programmes for salmon, seabass, seabream and rainbow trout can respond to future demands for novel feed compositions, and to make further improvements to disease resistance, climate resilience and animal welfare. For Atlantic salmon (Salmo salar) we estimate the genotype by diet/climate interactions (GxE) in semi-commercial salmon production system and validate best selection methods in salmon breeding programs by comparing traditional BLUP selection with GS/MAS for production and robustness traits. Estimated correlations will be taken as indicators of the magnitude of re-ranking of genotypes across diets/environments. Power calculations are elementary part of the experimental design but unfortunately often de-prioritized, compromising the critical interpretation of the results. A priori power calculations of genetic studies are characterized with multiple uncertainties, such as true relationship structure, number of families and individuals at the time of registration and unknown heritability of the traits of interest. This said, we argue that by performing a range of power calculations it is possible to frame the true power of the experiment and improve the probability of executing scientifically solid experiments with given restrictions of resources. We demonstrate the optimisation of the experimental design in order to have adequate power to detect significant GxE (diet/climate aggregate) given FutureEUAqua WP1 resources.