Avaliação genética da produção de leite no dia do controle de caprinos de aptidão leiteira utilizando modelos de regressão aleatória multicaracterísticos

• Ano de defesa: 2014
• Autor(a) principal: Silva, Felipe Gomes da
• Orientador(a): Torres, Robledo de Almeida
• Banca de defesa: Ventura, Henrique Torres , Nascimento, Moysés , Carneiro, Antônio Policarpo Souza
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Viçosa
• Programa de Pós-Graduação: Doutorado em Genética e Melhoramento
• Departamento: Genética animal; Genética molecular e de microrganismos; Genética quantitativa; Genética vegetal; Me
• País: BR
• Palavras-chave em Português: Melhoramento genético; Leite - Produção; Modelos de regressão aleatória
• Palavras-chave em Inglês: Genetic improvement; Milk - Production; Random regression models
• Área do conhecimento CNPq: CNPQ::CIENCIAS AGRARIAS::ZOOTECNIA::GENETICA E MELHORAMENTO DOS ANIMAIS DOMESTICOS
• Link de acesso: http://locus.ufv.br/handle/123456789/1408

The main objective of the present study was to conduct a genetic evaluation of test-day milk yield (TDMY) using multi-trait random regression models (MRRM). The specific objectives were: to study the genetic divergences between the first four lactations of Alpine goats using single-trait random regression models (SRRM) for each lactation; to demonstrate the application of the RENPED software to recode alphanumeric pedigrees; to determine different alternatives in the structures of truncation point, fixed effects and units of measurement of the TDMY response variable on the classification of animals evaluated genetically; and to determine the best random regression model to evaluate TDMY for Saanen goats. For this purpose, data from the dairy goat herd of Universidade Federal de Viçosa were used. The WOMBAT software was employed in all genetic evaluations; the CALC software from LibreOffice was used in datasets editions; and the RENPED software was used in the recoding of pedigree and datasets and a few descriptive analyses. Differences between the genetic structures of each one of the lactations should be studied to properly design selection strategies in dairy herds. Future analyses testing different models for the second, third and fourth lactations should be carried out. Analyses using the MRRM were feasible, but showed greater difficulty to converge than the separate SRRM analyses for each lactation. MRRM and SRRM show similar estimates for the variance components over the lactation curve. MRRM can provide greater advantages in relation to SRRM as they better utilize the data of at least two lactations simultaneously. RENPED is an alternative to correct errors and recode alphanumeric data as well as to make pedigree files and data available for analyses on various programs. There is a large difference in the classification of animals when the truncation point is too low in relation to medium and high truncation points for lactation. Better results are obtained in models that utilize the fixed year-season classificatory effect of the milk control through the construction of these fixed effects by clustering seasons that start in one year and end in another instead of only mixing the year and season variables. The use of TDMY in liters as traditionally adopted does not undermine the genetic evaluations with problems of numerical errors, but too large or too small units of measurement may generate this type of problem. The most parsimonious model to evaluate TDMY for Saanen goats was the one which considered a fixed regression of order 3, additive genetic effect of order 2, permanent individual effect of order 8, in addition to five classes of residual variance. It is also concluded that same-order combinations for the additive genetic and permanent individual curves may not find the best model according to the traditional comparison criteria.