4.2. Genetic analysis
The animal model containing common environmental effects used in this study proved to be an optimal model to estimate turbot genetic parameters (Guan et al., 2016; Ma et al., 2009). The fixed effect of sex was nonsignificant. This was in agreement with previous research. Although female turbot had higher body weight than males, the difference was not significant in the early stage (Wang et al., 2014). The day-age was the same in all individuals, therefore, the day-age covariant was omitted from the model. Heritability estimates for BW and BL obtained from both methods were all medium in 15 month-old turbot, based on the following categorization: low (0.05–0.15), medium (0.20–0.40), high (0.45–0.60), and very high (N0.65) (Cardellino and Rovira, 1987). A limited number of studies have reported on the estimation of growth trait heritability in turbot. Ma et al. (2008, 2009) estimated heritability for growth traits in turbot, they reported 0.450–0.514 for BW and 0.251–0.425 for BL in 6- month-old fish, and 0.34 for BW in 15-month-old fish. Both Liu et al. (2011) and Guan et al. (2016) estimated heritability for BW in 100 juvenile turbot and obtained 0.22 and 0.19, respectively. The results from both methods in our study were close to those reported in earlier studies. What calls for attention is that selecting 40 largest body weigh individuals from each family at 3 months post hatching induced a slight bias to the estimation of heritability potentially, because it might increase resemblance of the family members. Heritability and standard error based on molecular relatedness were slightly lower than pedigree, because molecular relatedness estimated a smaller genetic variance.
