Confiabilidade do valor genético predito em diferentes populações

• Ano de defesa: 2018
• Autor(a) principal: Feltes, Giovani Luis
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• dARK ID: ark:/26339/0013000003h15
• Idioma: por
• Instituição de defesa: Universidade Federal de Santa Maria; Brasil; Zootecnia; UFSM; Programa de Pós-Graduação em Zootecnia; Centro de Ciências Rurais
• Palavras-chave em Português: Confiabilidade; Correlação de Spearman; Herdabilidade; Regressão aleatória; Heritability; Reliability; Random regression models; Spearman correlation; CNPQ::CIENCIAS AGRARIAS::ZOOTECNIA
• Link de acesso: http://repositorio.ufsm.br/handle/1/14078

The purpose of this study was to evaluate the effect of reducing the number of lactation controls on the accuracy of the prediction of the genetic value of the animals. The database provided by the Brazilian Association of Breeders of the Dutch Breed was edited to contain only performance observations collected from cows with 10 monthly controls in lactation and was composed of 111,840 records of monthly production collected between 1991 and 2010 and from this, by random elimination of 1, 2, 3, 4, 5 and 6 controls, six other files were generated, totaling seven different populations, bringing together cows with 10, 9, 8, 7, 6, 5 and 4 controls in lactation, respectively. Genetic parameters and values were estimated by means of a random regression model with a polynomial of order 4 for genetic and permanent environment effects and homogeneous residual variances. Breeders were grouped according to the number of daughters to verify differences in mean reliability of predicted genetic values and Spearman rank correlation between genetic values of the best 10 and 20% breeding herds was calculated in the different populations to determine if there were changes in the ranking. Populations with number of partial controls had lower genetic variance, and consequently lower heritability for milk production; the mean reliability was lower in populations with lower number of controls; the number of bulls with high reliability was also lower; the Spearman correlations presented mean values, indicating that there are changes in the ranking of the best animals. Therefore, the reduction in the number of lactation controls negatively affects the genetic evaluation of populations.