Genetic analysis of production traits in turbot (Scophthalmus maximus) using random regression models based on molecular relatedness

Abstract

AbstractReliable estimates of genetic parameters for growth traits as a trajectory of age are needed to optimize existing turbot breeding programmes. To evaluate the potential of early selection strategies, the use of biometric body measurements, length (L), width (W) and area (A), at early ages as alternative indicators for the selection trait at harvest was explored. Random regression model (RRM) based on molecular relatedness (MR) was used to analyse the trajectory of genetic parameters for growth traits in turbot from 162 to 756 days posthatch (dph). Heritability estimates for body weight (BW) ranged from 0.34 to 0.54. Heritability estimates for W, A and L were also moderate to high ranging from 0.18 to 0.43. Estimates for L and W declined with age, while those for A increased towards harvest age. Genetic (rG) and phenotypic (rP) correlations between BW and the three morphometric traits L, A and W were estimated using simple bivariate animal models at young (AC1), medium (AC2) and old (AC3) age classes. Correlations between BW and morphometric body traits were high, ranging from 0.7 to 0.9 in all three age groups. Genetic correlations between traits were highest (>0.9) in AC3. To explore the potential for early selection, genetic correlations were derived from the RRM between all days of measurement for all traits separately. From dph 300 onwards, intratrait estimates of rG were moderate to high (above 0.7 for dph 410 and higher ages for traits BW, L and A). Results showed that genetic selection for BW, L and A is promising and that A and L could be successfully used as alternative indicator traits if measurements of BW are not available. Large BW and A at harvest could be achieved as a correlated response to early selection for these traits at around 500 dph.