Application of mixed-model techniques to fish breed improvement: analysis of breeding-value selection to increase 98-day body weight in tilapia

摘要

A post-selection analysis of a three-generation selection experiment was undertaken to assess the effectiveness of using estimated breeding values as the selection criterion for the improvement of body weight at 98 days of age. Two replicate selection lines and a control line were derived from a common base population. Fixed effects of generation, line, sex and tank, as well as the fixed-effect interaction of the generation with line, sex and tank, were the significant factors determining performance. The estimated heritability of the 98-day weight was 0.20 +/- 0.04, while the maternal direct-genetic and full-sib group effects as a proportion of phenotypic variance were 0.02 +/- 0.02 and 0.09 +/- 0.02, respectively. Although phenotypic means varied from generation to generation, the genetic gains calculated from the estimated breeding values were 2.61 +/- 0.05 and 2.42 +/- 0.37 g per generation for the replicate lines, which is equivalent to a 40% improvement in just three generations. Additionally, although the randomly selected control line showed a linear increase in performance of 4.10 +/- 1.27 g per generation, the change in the average breeding value was only 0.27 +/- 0.01 g per generation, indicating that the increase in performance was due to improved management and/or adaptation to the culture environment. The post-selection analysis clearly demonstrated the superiority of the mixed-model approach to selection, with the potential of increasing the selection response by 20-30% above the value expected from simple mass selection.