Main content area

Genetic parameters of economically important traits recorded at a given age in striped catfish (Pangasianodon hypophthalmus)

Sang, Nguyen Van, Klemetsdal, Gunnar, Ødegård, Jørgen, Gjøen, Hans Magnus
Aquaculture 2012 v.344-349 pp. 82-89
Pangasianodon hypophthalmus, additive effect, animal models, body size, body weight, carcass characteristics, color, equations, fillets, genetic correlation, heritability, lipids, model validation, prediction, sires, spawning, statistical models, variance
The objective of this study was to estimate heritabilities of and genetic and phenotypic correlations between several body size and carcass quality traits in striped catfish (Pangasianodon hypophthalmus), when recorded at about the same age. Predicted data were obtained for fillet weight, fillet yield and fillet fat based on a prediction equation. Partial factorial and nested mating designs were used, respectively, to generate the G2 and G3 generations, with 162 and 156 full-sib families (from 95 sires and 97 dams, and 93 sires and 156 dams in the two generations, respectively). The data-set in the G2 generation consisted of 2767 records (2567 records for predicted fillet weight and predicted fillet yield, as 200 records were used to construct the prediction equations), while 5640 records (body weight and standard length) were available in the G3 generation. Model validation by model fitting (for random effects) and predictive ability (for fixed effects) concluded the statistical model used to analyse the data; that should contain random effects of additive genetic and common environment, and fixed covariates of number of days from spawning till tagging and number of days from tagging till first harvest, respectively. A univariate animal model was used to estimate variance components, while a bivariate model was used to estimate genetic correlations. Moderate heritability estimates were obtained for body weight (0.21–0.34) and fillet weight (0.19–0.22), while low heritability estimates were found for fillet yield (0.03–0.05), intestinal fat (0.04), predicted fillet fat percentage (0.04) and fillet colour (0.04). Of predicted traits, only that for fillet weight showed a heritability with the potential of utilising the information on the selection candidate itself (h²=0.10). As most of the variation of fillet weight is explained by the variation in body weight and the two traits are highly genetically correlated (0.95–0.96), this allows for efficient indirect selection for fillet weight through direct selection for body weight. Fillet weight (and body weight) had a moderate positive genetic correlation with fillet fat and was almost uncorrelated with fillet colour, while the two latter traits were also genetically uncorrelated. Thus, selection for fillet weight and/or body weight is expected to increase the fillet fat percentage indirectly. With a goal of keeping the fat percentage of the fillets unchanged, both traits groups need to be considered simultaneously in the breeding program.