Interação genótipo X ambiente para peso vivo e modelagem estatística para seleção genética em tilápias do nilo (Oreochromis niloticus)

• Autor(a) principal: Oliveira, Sheila Nogueira de
• Data de Publicação: 2013
• Tipo de documento: Tese
• Idioma: por
• Título da fonte: Repositório Institucional da Universidade Estadual de Maringá (RI-UEM)
• Texto Completo: http://repositorio.uem.br:8080/jspui/handle/1/1578
• topic: Tilápia do Nilo; (Oreochromis niloticus); Melhoramento genético; Linhagem GIFT; Modelagem de dados; Genética quantitativa animal; Inferência Bayesiana; Interação genótipo-ambiente; Paraná; Brasil.; Nile tilapia; (Oreochromis niloticus); Genetic improvement; GIFT strain; Data modeling; Animal quantitative genetics; Bayesian Inference; Genotype-environment interaction; Paraná; Brazil.; Ciências Agrárias; Zootecnia

This study evaluated the existence of genotype x environment interaction for bodyweight in Nile tilapia (Oreochromis niloticus), GIFT range between three cities indifferent regions in the Paraná state. The data set consisted of 1,132 animals, males and females, born between November 2011 and February 2012. Analyses were performed using Bayesian inference, considering an animal model that included fixed effects of sex, linear and quadratic covariate of fish age (in days), in addition to the effects considered random, and the additive genetic and common environment hatchery (c) and nursery (w) , where the weight was treated as a different trait in each of the regions. The results of heritability were high for univariate models, being 0.71 , 0.72 and 0.67 for the cities of Palotina (PL) , Floriano (FL) and Diamante do Norte (DN) , respectively , the results of heritability for the two-trait analyzes showed similar values. Genetic correlations estimated in vicariate analyzes were weak with values between 0.12 PL-FL,0.06 for PL- DN and 0.23 for FL-DM. The Spearman correlation values were low, indicating change in ranking in the animals selection in different environments understudy. There was heterogeneity of phenotypic variance between the three regions and heterogeneity of residual variance between PL and DN. The direct genetic gain was greater for DN region with the value of 281.35 g of weight gain per generation, followed by FL with 198.24 g per generation and finally PL with 98.73 g per generation. The indirect genetic gains ranged from 7.77 g per generation between DNand PL to 74.66 g per generation between FL and DN. Eight statistical models were used to check what best describes the average daily weight gain (ADG), in order to make the best selection in animals subjected to genetic breeding program of tilapia(Oreochromis niloticus), variety GIFT, and the data set consisted of information from2,615 animals, from the fourth generation of selection (G4) . The analyzes considered the animal model that included fixed effects of sex, linear and quadratic covariate offish age, in days, in addition to additive genetic effects. The models were modified to include or not maternal genetic effect (m) and common hatchery environments (c) and nursery (w) . The results of heritability were considered the average for most models,M2 = 0.24, M4 = 0.23, M5 = 0.28, = 0.30 M6, M7 and M8 = 0.33 = 0.36 and high for models M1 = M3 = 0.83 and 0.79 . The Bayesian deviance information criterion (DIC)was lower for M1 (DIC = -282.59) and higher for M4 (DIC = 1754.57) . The criterion of log marginal density for Bayes factor agreed with the DIC to the lowest value at which M1 = 585.29 and the highest value presented for this criterion was for M5 =1837.37.