Longitudinal genetic analyses of fillet traits in Nile tilapia Oreochromis niloticus

Abstract There is little published information on genetic covariance estimates among meat yield traits in Nile tilapia. Longitudinal genetic analyses to evaluate patterns of environmental and genetic covariances among such traits during growth would help to establish selection criteria and predict direct and correlated response to selection. Accordingly, longitudinal genetic studies from 106 to 245 days of age were performed by random regression models for carcass and fillet weights and yields in Nile tilapia. For each analysis, the statistical model included random family common environmental and genetic effects and assuming heterogeneity of residual variances. For carcass and fillet yield, genetic correlations between closely paired ages (from 140 to 200 days) were higher (> 0.80) than those separated by moderate (> 0.40) or longer (approaching zero) intervals. For carcass and fillet weight, genetic correlations when both ages were greater than 125 days post-hatching were larger than 0.60, and those when both ages were greater than 149 days post-hatching were larger than 0.80. Heritability estimates for fillet and carcass yields ranged from moderate (0.12) to high (0.52), suggesting the likelihood that selection programs could improve these traits. The manner in which fillets were removed seems to influence the results. Heritability estimates for carcass and fillet weight ranged from high (0.52) to low (0.01). The family common environmental effect was important for reducing the heritability estimates of these two traits, probably due to high correlation of each trait to body weight at slaughter, which is substantially affected by the common family environmental effect. Accordingly, to increase the accuracy of predicted breeding values for these traits, it is important to reduce family effects by minimizing the duration during which families are reared separately, then tagging individual fish and initiating communal stocking at as early a life stage as possible. Based upon estimated genetic parameters among the traits, direct selection for body weight, carcass weight and fillet weight is not recommended if the breeding goal is to increase fillet yield as a correlated response. Contrarily, the high genetic correlations between body weight and fillet and carcass weights (> 0.98) suggest that body weight could be used as a selection criterion to avoid slaughtering potential breeders to acquire data and to achieve higher selection intensity.