Modelos para estimação de componentes de (co)variância para pesos do nascimento à idade adulta de animais da raça Nelore
Carregando...
Data
2011-08-09
Autores
Boligon, Arione Augusti [UNESP]
Título da Revista
ISSN da Revista
Título de Volume
Editor
Universidade Estadual Paulista (Unesp)
Resumo
O presente estudo foi desenvolvido com o objetivo de estudar a variação genética de pesos obtidos do nascimento à idade adulta, utilizando modelos multi-características (MC) e de regressão aleatória (MRA), visando obter modelos adequados e parcimoniosos para a estimação de componentes de (co)variância e predição de valores genéticos de animais da raça Nelore, os quais possam ser utilizados em programas de melhoramento genético. As medidas de peso utilizadas são provenientes de oito fazendas participantes do Programa de Melhoramento Genético da Raça Nelore (PMGRN). As análises MC foram realizadas considerando nove pesos: peso ao nascer, desmama, ano, sobreano e aos 2, 3, 4, 5 e 6 anos de idade. Foram estudados os seguintes modelos: multi-características padrão, cinco modelos de posto reduzido ajustando os primeiros 1, 2, 3, 4 e 5 componentes principais genéticos e cinco modelos utilizando análise de fatores com 1, 2, 3, 4 e 5 fatores. Em todos os modelos foram considerados os efeitos aleatórios genético aditivo direto e residual. Para o peso ao nascer e à desmama também foram incluídos, como aleatórios, os efeitos genético materno e ambiente permanente materno. Foram considerados os efeitos fixos de grupo de contemporâneos e efeitos linear e quadrático da idade do animal à pesagem (exceto para o peso ao nascer) e idade da mãe ao parto. As matrizes de (co)variâncias genética materna, de ambiente permanente materno e residual foram assumidas terem posto completo. Os resultados indicam que apenas três componentes principais são requeridos para modelar a estrutura das (co)variâncias genéticas entre pesos do nascimento à idade adulta, reduzindo assim o número de parâmetros nas análises. Nas análises utilizando MRA foram utilizados pesos de fêmeas do nascimento...
The genetic variation of weights obtained from birth to adult ages were studied using multi-trait (MT) and random regression (RRM) models in order to obtain adequate and parsimonious models for the estimation of (co)variance components and prediction of genetic values of Nellore cattle, which can be used in breeding programs. The weights traits were obtained from eight herds participating in the Nelore Cattle Breeding Program (NCBP). In the MT analyses were included weight records taken at nine standard ages (at birth, weaning, 365 and 550 days of age and 2, 3, 4, 5 and 6 years of age). A standard multivariate analysis, reduced rank analyses fitting the first 1, 2, 3, 4 and 5 genetic principal components, and analyses that fitted a factor analytic structure considering 1, 2, 3, 4 and 5 factors, were carried out. All models included genetic additive direct and residual as random effect. For birth and weaning weight also was considered genetic maternal and maternal permanent environmental effects as random. Contemporary groups, age at weighing (except for birth weight) and dam age at calving was included as fixed effect. The genetic maternal, maternal permanent environmental and residual covariance matrices were assumed to have full rank throughout. The results indicate that only three principal components are required to model the genetic covariance structure among weight from birth to mature, decreasing the number of parameter in the analyses. For RRM were used female weights from birth to mature. The models included direct additive and maternal genetic effects and animal and maternal permanent environmental effects as random. Contemporary group and dam age at calving within the age class of the animal (linear and quadratic effect) were included as fixed effects, and orthogonal Legendre... (Complete abstract click electronic access below)
The genetic variation of weights obtained from birth to adult ages were studied using multi-trait (MT) and random regression (RRM) models in order to obtain adequate and parsimonious models for the estimation of (co)variance components and prediction of genetic values of Nellore cattle, which can be used in breeding programs. The weights traits were obtained from eight herds participating in the Nelore Cattle Breeding Program (NCBP). In the MT analyses were included weight records taken at nine standard ages (at birth, weaning, 365 and 550 days of age and 2, 3, 4, 5 and 6 years of age). A standard multivariate analysis, reduced rank analyses fitting the first 1, 2, 3, 4 and 5 genetic principal components, and analyses that fitted a factor analytic structure considering 1, 2, 3, 4 and 5 factors, were carried out. All models included genetic additive direct and residual as random effect. For birth and weaning weight also was considered genetic maternal and maternal permanent environmental effects as random. Contemporary groups, age at weighing (except for birth weight) and dam age at calving was included as fixed effect. The genetic maternal, maternal permanent environmental and residual covariance matrices were assumed to have full rank throughout. The results indicate that only three principal components are required to model the genetic covariance structure among weight from birth to mature, decreasing the number of parameter in the analyses. For RRM were used female weights from birth to mature. The models included direct additive and maternal genetic effects and animal and maternal permanent environmental effects as random. Contemporary group and dam age at calving within the age class of the animal (linear and quadratic effect) were included as fixed effects, and orthogonal Legendre... (Complete abstract click electronic access below)
Descrição
Palavras-chave
Nelore (Zebu) - Peso, Nelore - Mature weight
Como citar
BOLIGON, Arione Augusti. Modelos para estimação de componentes de (co)variância para pesos do nascimento à idade adulta de animais da raça Nelore. 2011. xii, 97 f. Tese (doutorado) - Universidade Estadual Paulista, Faculdade de Ciências Agrárias e Veterinárias, 2011.