The direct relationship between myostatin (MSTN) and muscle growth has led many researchers to study, with the aim of genetic improvement, the effect of MSTN gene on performance traits in producing species, such as pigs, sheep, horses, broiler chickens and in some species of fish. However, the MSTN gene effect has not yet been measured in genetic prediction models for neotropical fish. We verified the effect of incorporating molecular information from the MSTN gene on genetic prediction models to estimate heritability and breeding values for pacu weight.
A performance test was carried out with 292 pacus, from 14 full-sib families, in ponds until the animals reached an average weight of 711.27 ± 283.45 g. After the test, the animals were genotyped, by molecular techniques, for a STR marker positioned in the 3’UTR region of the MSTN gene. With the data, two animal models were performed, one univariate considering only the weight of the animals as the analysis variable, and another bivariate (incorporation model), where we include the weight and the effect of each animal’s genotype as analysis variables. The genotypic effect was calculated as the difference between the average weight of the genotypes and the average overall weight of the animals. With both models, we estimated the heritability and the breeding values (EBVs) of animals for weight. The EBVs were used to rank the animals, and the ranks were used to verify, by correlation analysis, the change in the positioning of the animals when molecular information was incorporated in the model.
The incorporation of molecular information into the genetic prediction model generated a larger and much more accurate estimate of heritability for weight, which suggests a more efficient genetic selection process using the bivariate model (Table 1). The estimate of additive genetic variance was also more accurate due to the incorporation model. The correlation between the ranking of the EBVs of the animals considering the two models was less than 0.6, which indicates an important change in the positioning of the animals in the ranking when the prediction model is changed (Table 1).