Avaliação genética de crescimento e resistência a verminoses em ovinos Santa Inês utilizando modelos de regressão aleatória e análises multivariadas
Carregando...
Arquivos
Data
2018-02-21
Autores
Freitas, Luara Afonso de
Título da Revista
ISSN da Revista
Título de Volume
Editor
Universidade Estadual Paulista (Unesp)
Resumo
O uso de raças ou animais resistentes aos endoparasitas dentro do sistema de produção de ovinos de carne é uma das alternativas mais eficientes para se obter maior produção. O controle dos acasalamentos entre animais aparentados é de grande importância para reduzir perdas na produção pelo aumento da homozigose de alelos deletérios. O peso corporal é uma das características de maior importância econômica em ovinos e pode ser registrado ao longo da vida de um mesmo animal. Os modelos de regressão aleatória na avaliação genética de medidas longitudinais são geralmente recomendados para estimar parâmetros genéticos de dados longitudinais. As análises multivariadas são um conjunto de métodos estatísticos que torna possível a análise simultânea de medidas múltiplas para cada indivíduo ou fenômeno observado. Este trabalho teve como objetivos: (1) analisar a estrutura populacional e verificar o efeito da endogamia sobre características fenotípicas de crescimento e resistência a verminose, (2) comparar estruturas (homogeneidade e heterogeneidade de variâncias) para modelar a variância residual em modelos de regressão aleatória, (3) avaliar a influência desta modelagem na estimação de parâmetros genéticos a fim de determinar o modelo mais apropriado para estudar o crescimento e a resistência à verminose e (4) avaliar a associação genética e explorar o perfil genético da produção de carne e resistência à verminose em ovinos da raça Santa Inês. O banco de dados fenotípico conteve 2.159 registros de peso corporal oriundos de 753 animais, 1.468 registros de volume globular provenientes de 476 animais e 981 registros de contagem de ovos por grama de fezes oriundos de 375 animais entre 30 a 900 dias de idade medidos mensalmente e registros de pedigree de 2.410 animais da raça Santa Inês. Os animais foram divididos em 10 classes de idade (CI90 = 30 a 90 dias de idade CI180= 91 a 180 dias de idade, CI270= 181 a 270 dias de idade, CI360= 271 a 360 dias de idade, CI450= 361 a 450 dias de idade, CI540= 451 a 540 dias de idade, CI630= 541 a 630 dias de idade, CI720= 631 a 720 dias de idade, CI810= 721 a 810 dias de idade, CI900= 811 a 900 dias de idade). A estrutura do pedigree foi analisada utilizando o programa Endog 4.8. As análises de estimação dos componentes de variância foram realizadas por meio de modelo animal unicaracterística de regressão aleatória pelo método de máxima verossimilhança restrita utilizando polinômios de Legendre. O critério da informação bayesiana foi utilizado para selecionar o melhor modelo. A análise de componentes principais foi realizado no software STATISTIC e a analise de agrupamento hierárquico e não hierárquico no software R. Foi observado aumento no coeficiente de endogamia até a 6ª geração máxima, aumento do parentesco médio e da porcentagem de animais endogâmicos até a 7ª geração máxima. O polinômio de Legendre de ordem 3 (intercepto, linear e quadrático) proporcionou o melhor ajuste dos dados para peso corporal, volume globular e contagem de ovos por grama de fezes. A seleção dos animais poderia ser feita com base no peso corporal a partir de 271 dias de idade, pois a estimativa de herdabilidade foi de 0,23 (0,03) e foram altas as correlações genéticas do peso corporal nesta idade com a maioria dos pesos corporais nas demais idades, exceto com peso corporal medida entre 30 a 90 dias de idade. A seleção destes animais para resistência a verminose se realizada poderá ser pouco eficiente devido às baixas estimativas de herdabilidade para volume globular (0,09±0,01 a 0,18±0,02) e contagem de ovos por grama de fezes (0,01±0,02 a 0,17±0,03), portanto, o ganho genético será a longo prazo. A seleção para peso corporal poderia trazer ganhos genético favoráveis para a característica volume globular devido à proximidade das características no eixo do componente principal um. Nas análises de agrupamento apenas um dos três grupos genéticos identificados apresentou perfil genético indicado para a seleção para peso corporal e volume globular.
The use of resistant animals to gastrointestinal nematode parasites within the meat production system is one of the most efficient alternatives to increase meat yield in sheep. The control of mating between related animals is of great importance to reduce losses in meat production by increasing the homozygosity of deleterious alleles. Application of Random regression models in the genetic evaluation of longitudinal measurements is generally recommended to adjust growth curves and to estimate genetic parameters of longitudinal data. The multivariate analyzes are a set of statistical methods that make possible the simultaneous analysis of multiple measures for each individual or observed phenomena. The aims of this study were: (1) to analyze the population structure and verify the effect of inbreeding on phenotypic traits of growth and resistance to gastrointestinal nematode parasites, (2) to compare structures (homogeneity and heterogeneity of variances) to model residual variance in random regression models, (3) to study the influence of this modeling in the estimation of genetic parameters in order to determine the most appropriate model to study the growth and resistance to gastrointestinal nematode parasites curves, and (4) evaluate the genetic association and explore the genetic profile of meat production and resistance to gastrointestinal nematode parasites in Santa Inês sheep. The data contained 2,159 body weight records from 753 animals, 1,468 mean corpuscular volume records from 476 animals and 981 records of fecal egg counts from 375 animals and pedigree information from 2,410 in Santa Ines sheep. The animals were divided into 10 age classes (CI90 = 30 to 90 days of age CI180 = 91 to 180 days of age, CI270 = 181 to 270 days of age, CI360 = 271 to 360 days of age, CI450 = 361 to 450, CI540 = 451 to 540 days of age, CI630 = 541 to 630 days of age, CI720 = 631 to 720 days of age, CI810 = 721 to 810 days of age, CI900 = 811 to 900 days of age). The structure of the pedigree was analyzed using the program Endog 4.8. Analyzes of the variance component estimates were performed using a one-trait model of random regression using the restricted maximum likelihood method. The Bayesian information criterion was used to select the best model. The principal component analysis was performed in STATISTIC software and hierarchical and non-hierarchical analysis grouping was performed in the software R. There was an increase in the inbreeding coefficient up to the 6th maximum generation and also an increase in the average relatedness and in the percentage of inbred animals up to the 7th maximum generation. Genetic correlation estimates for all traits in the age classes presented values close to the unit at close weighings, with a reduction in the magnitude of the correlations as the time interval between weighings increased. The Legendre polynomial of order 3 (intercept, linear and quadratic) provided the best fit of the data for body weight, mean corpuscular volume, and fecal egg counts. The selection of the animals could be made based on body weight from 271 days of age, as the heritability estimate was moderate, 0.23 (0.03), and showed high genetic correlations with most body weights, except with body weight from 30 to 90 days of age. Selection of these animals for resistance to gastrointestinal nematode parasites if carried out may result in low efficiency due to the low estimates of heritability for the mean corpuscular volume (0.09 ± 0.01 to 0.18 ± 0.02) and fecal egg counts (0.01 ± 0.02 to 0.17 ± 0.03), therefore, the genetic gain will be low and in the long term. The selection of body weight could bring favorable genetic gains to the mean corpuscular volume due to the proximity of the traits in the axis of the main component one. In cluster analysis, only one of the three genetic groups identified has a genetic profile indicated for selection for the studied traits.
The use of resistant animals to gastrointestinal nematode parasites within the meat production system is one of the most efficient alternatives to increase meat yield in sheep. The control of mating between related animals is of great importance to reduce losses in meat production by increasing the homozygosity of deleterious alleles. Application of Random regression models in the genetic evaluation of longitudinal measurements is generally recommended to adjust growth curves and to estimate genetic parameters of longitudinal data. The multivariate analyzes are a set of statistical methods that make possible the simultaneous analysis of multiple measures for each individual or observed phenomena. The aims of this study were: (1) to analyze the population structure and verify the effect of inbreeding on phenotypic traits of growth and resistance to gastrointestinal nematode parasites, (2) to compare structures (homogeneity and heterogeneity of variances) to model residual variance in random regression models, (3) to study the influence of this modeling in the estimation of genetic parameters in order to determine the most appropriate model to study the growth and resistance to gastrointestinal nematode parasites curves, and (4) evaluate the genetic association and explore the genetic profile of meat production and resistance to gastrointestinal nematode parasites in Santa Inês sheep. The data contained 2,159 body weight records from 753 animals, 1,468 mean corpuscular volume records from 476 animals and 981 records of fecal egg counts from 375 animals and pedigree information from 2,410 in Santa Ines sheep. The animals were divided into 10 age classes (CI90 = 30 to 90 days of age CI180 = 91 to 180 days of age, CI270 = 181 to 270 days of age, CI360 = 271 to 360 days of age, CI450 = 361 to 450, CI540 = 451 to 540 days of age, CI630 = 541 to 630 days of age, CI720 = 631 to 720 days of age, CI810 = 721 to 810 days of age, CI900 = 811 to 900 days of age). The structure of the pedigree was analyzed using the program Endog 4.8. Analyzes of the variance component estimates were performed using a one-trait model of random regression using the restricted maximum likelihood method. The Bayesian information criterion was used to select the best model. The principal component analysis was performed in STATISTIC software and hierarchical and non-hierarchical analysis grouping was performed in the software R. There was an increase in the inbreeding coefficient up to the 6th maximum generation and also an increase in the average relatedness and in the percentage of inbred animals up to the 7th maximum generation. Genetic correlation estimates for all traits in the age classes presented values close to the unit at close weighings, with a reduction in the magnitude of the correlations as the time interval between weighings increased. The Legendre polynomial of order 3 (intercept, linear and quadratic) provided the best fit of the data for body weight, mean corpuscular volume, and fecal egg counts. The selection of the animals could be made based on body weight from 271 days of age, as the heritability estimate was moderate, 0.23 (0.03), and showed high genetic correlations with most body weights, except with body weight from 30 to 90 days of age. Selection of these animals for resistance to gastrointestinal nematode parasites if carried out may result in low efficiency due to the low estimates of heritability for the mean corpuscular volume (0.09 ± 0.01 to 0.18 ± 0.02) and fecal egg counts (0.01 ± 0.02 to 0.17 ± 0.03), therefore, the genetic gain will be low and in the long term. The selection of body weight could bring favorable genetic gains to the mean corpuscular volume due to the proximity of the traits in the axis of the main component one. In cluster analysis, only one of the three genetic groups identified has a genetic profile indicated for selection for the studied traits.
Descrição
Palavras-chave
AIREML, componentes principais, ovinos deslanados, Ovis aries, hair sheep, principal components