Avaliação do crescimento e de modelos de regressão aleatória em caprinos da raça Alpina

• Ano de defesa: 2011
• Autor(a) principal: Oliveira, Joashllenny Alves de
• Orientador(a): Euclydes, Ricardo Frederico
• Banca de defesa: Campelo, José Elivalto Guimarães
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Viçosa
• Programa de Pós-Graduação: Mestrado em Zootecnia
• Departamento: Genética e Melhoramento de Animais Domésticos; Nutrição e Alimentação Animal; Pastagens e Forragicul
• País: BR
• Palavras-chave em Português: Endogamia; Curva de crescimento; Parda Alpina; Anglo-nubiana
• Palavras-chave em Inglês: Endogamy; Growth curve; Alpina brown; Anglo-nubian
• Área do conhecimento CNPq: CNPQ::CIENCIAS AGRARIAS::ZOOTECNIA::GENETICA E MELHORAMENTO DOS ANIMAIS DOMESTICOS
• Link de acesso: http://locus.ufv.br/handle/123456789/5685

It was used 1,501 records of body weights from birth to 196 days of age from 118 males and 103 females, in a total of 221 Alpina goats breed, from State Company for Agricultural Research in Paraiba (EMEPA - PB), collected between 2002 to 2009. The first analysis was to evaluate different non-linear models (Brody, Gompertz, Logistic, Richards and von Bertalanffy) that better describe the average growth curve and analyze the influence of environmental effects on the estimated parameters. The parameters: A is asymptotic value; B is integration constant; k is maturing rate; and m is inflection point, were estimated using NLIN procedure of SAS. By the goodness of fit criteria, as residual mean square (QMR), coefficient of determination (R2), convergence percentage (C%) and absolute mean error (DMA), it was found that the Gompertz model had higher goodness of fit then the other models to describe the growth curve of Alpina goats breed. The parameters A and k estimated by the Gompertz model were 25.2136 kg and 0.0163, respectively. It was found a significant reduction of absolute growth rate in the post-weaning period, and should be adopted adequate dietthe animal nutritional requirements in this period. The effects of environment influenced significantly (P <0.05) the parameters A and k. The correlation between estimated parameters and k was negative (-0.5435), indicating that the earliest animals have less probability to reach higher weights at adult age. For the second analysis of these dataset, it was compared different random regression models, adjusted by different orders Legendre polynomials, and the inclusion of additive genetic direct, maternal additive genetic, animal permanent environmental and maternal permanent environmental effects, to evaluate the best model to fit the growth curve of these animals in the study. Adjust orders Legendre polynomials of 3 (quadratic function) and 4 (cubic function) were used to describe the structures of (co) variance function in time from random regression models, compared by the maximum likelihood logarithm (Log L), Akaike information criterion (AIC) and Schwarz Bayesian information (BIC). Thus, it was found the two best models and then they were compared by the LRT (likelihood ratio test) (P <0.01) to verify the possible difference between these models. WOMBAT program was used to estimate the components of (co)variances. The best models that describe the changes in (co) variance structure in the average growth curve was obtained by Legendre polynomial function with orders 3,4 (additive genetic effects and animal permanent environmental) and 3,3,4 (additive genetic effects, maternal additive genetic and animal permanent environment) in Alpina goats breed.