Estimativas de componentes de variâncias e de valores genéticos para características de crescimento, reprodução e habilidade materna em ovinos da raça Somalis Brasileira

• Ano de defesa: 2010
• Autor(a) principal: Magalhães, Ana Fabrícia Braga
• Orientador(a): Lôbo, Raimundo Nonato Braga
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Palavras-chave em Português: Zootecnia; Ovino - Crescimento; Genética animal; Ovino - Melhoramento genético; Análise multivariada
• Link de acesso: http://www.repositorio.ufc.br/handle/riufc/1183

The aims of this study were to estimate genetic parameters for growth, reproductive and maternal traits and assess the different adjustments of fixed effects in the prediction of breeding values for weight and weight gain at weaning of Brazilian Somali sheep, reared in semi-arid. Data from the Núcleo de Conservação da Embrapa Caprinos e Ovinos, located in the city of Sobral, north of Ceará state, suported by Programa de Melhoramento Genético de Caprinos e Ovinos de Corte (GENECOC). The analyzed growth traits were birth weight (PN), weaning weight (PD), weight gain from birth to weaning (GND), adjusted weight at 84 days (P84), weight gain from birth to 84 days (GN84) and adult weight (PA). The traits related to maternal ability were litter weight at birth (PTCN), litter weight at weaning (PTCD), mother's weight at weaning of the lambs (PW) and ratio of weaning (REL), which is represented by the rate between the total weight of lambs at weaning and the mother's metabolic weight. The analyzed reproductive traits were lambing interval (IP), lambing date (DP) and number of services per conception (NSC). The fixed models of analysis were defined by the MIXED procedure of SAS software, after checking the constraints and limitations of the data. (Co) variances and genetic parameters were estimated by Derivative Free Restricted Maximum Likelihood (DFREML) method using the MTDFREML software. Initially, univariate analyses were performed for each trait and then multivariate analyses were performed. An analysis of correlation of classification or "rank", using the Spearman coefficient, was performed to compare the rank of animals by PD, P84, GND and GN84, based on the additives genetic direct and maternal values. Heritabilities for growth traits estimated in univariate analysis were of low magnitude, ranging from 0.00 to 0.22. There was a trend of increasing of heritabilities in the multivariate analysis. In multivariate analysis for the growth traits, heritabilities for PN-PD-GND and PN-P84- GN84 ranged from 0.00 to 1.00. Genetic correlations also showed large variations, XVII from negative, zero to high and positive, ranging among PN-PD-GND from -1.00 to 0.94 and from -0.45 to 0 97 among PN-P84-GN84. The reproductive traits showed low heritabilities. The heritabilities for maternal traits were higher in the multivariate analysis, with moderates values and high correlation between PTCN and PTCD and unitary negative value between PW and REL. The weaning weight adjusted for the covariate age on the day of weighing and the weaning weight adjusted for standard age at weaning, in this study, 84 days, are traits of distinct natures under analyses. Using these criteria promotes different responses to selection, with different rank orders of animals. However, it was not possible to specify the criteria with better efficiency since the differences between the results were not fully understood