Interação genótipo x ambiente em codornas de corte e suas implicações em populações sob seleção

Abstract

The genotype x environment interaction is an important aspect that should be considered in the genetic evaluations of populations under selection. Reaction norms models are often used for the study of the genotype x environment interaction and can be applied to populations that have the phenotype of a group of related individuals measured along an environmental gradient. Many researchers assume that the reaction norm of a given genotype along the environmental gradient should be linear. However, imposing a linear model as more adequate to describe the behavior of a certain characteristic implies applying restrictions that prevent its genetic progress, that is, its evolution throughout the different environments. In general, in order to understand the evolution of the characteristic along the environmental gradient, the genetic parameters estimated through models of reaction norm must be analyzed in a complementary way. In this dissertation, we aimed to study the interaction genotype x environment in two lines of European quails in the final stage of growth created under different levels of methionine + cystine digestible apparent. For this, the weight gain was measured from 21 to 35 days old of the quails. The levels tested were 0.60; 0.70; 0.80; 0.90 and 1.00% methionine + cysteine digestible in the quails of the EV1 strain and 0.61; 0.71; 0.81; 0.91 and 1.01% methionine + cystine digestible in quails of the EV2 strain. Legendre polynomial ranging from one to five orders were used for the adjustment of fixed and random effects and heterogeneity of residual variance was considered. The models that fit best for each strain differed according to the order of the Legendre polynomial used to describe the mean trajectory. Four-order polynomials for the EV1 and first-order polynomials for the EV2 strain were required. In relation to the adjustment of the additive genetic effects, fourth-order polynomials were required for modeling both strains. Also, for both strains models that considered heterogeneity of residual variance with five classes showed best fit. After determining the model that best described the weight gain along the levels methionine + cystine digestible in both strains, genetic parameters were estimated. Correlation estimates and reaction norms of the genotypes as a function of levels of the environment gradient indicated interaction between genotype and the amino acid level supplied to the quails for weight gain in the final stage of growth. Heritability estimates ranged from 0.37 to 0.72 throughout the environments in EV1 strain and from 0.20 to 0.75 throughout the environments in EV2 strain. In addition, high values of coefficient of additive genetic variation were observed, ranging from 0.09 to 0.17 in the EV1 strain and 0.11 to 0.19 in the EV2 strain. The results indicate that there is genetic variability for weight gain in quails of both strains and that the characteristic in study can evolve over the generations through selection