Avaliação de métodos de controle de Endogamia utilizando dados simulados

• Autor(a) principal: Paiva, André Luis da Costa
• Data de Publicação: 2010
• Tipo de documento: Tese
• Idioma: por
• Título da fonte: LOCUS Repositório Institucional da UFV
• Texto Completo: http://locus.ufv.br/handle/123456789/1865
• topic: Endogamia; Gado holandês; Consaguinidade; Inbreeding; Dutch cattle; Consanguinity; CNPQ::CIENCIAS AGRARIAS::ZOOTECNIA::GENETICA E MELHORAMENTO DOS ANIMAIS DOMESTICOS

Inbreeding or consanguinity is a mating pattern that consists in bonding individuals with a certain level of relatedness. Different methods have been proposed to reduce the inbreeding rate in such a way that genetic gain is not affected. Data simulated through an up-to-date of the GENESYS 2010 program (EUCLYDES, 1996) were used in order to test the use of compensatory mating in simulated populations, the increase of the number of males selected as a tool for controlling inbreeding, controlling inbreeding in populations with different selection intensities for females, controlling inbreeding in populations selected based on the best linear unbiased prediction (BLUP) considering a higher heritability value. It was simulated a genome of 958 centimorgans in length, considering a trait whose expression was determined by 200 loci with two alleles per locus distributed across 40 chromosomes. Only one quantitative trait (slaughter weight) was evaluated with an initial mean phenotypic value of 2.3kg (5.07lb), standard deviation of 0.3kg (0.66lb) and heritability (h2) of 0.30. In order to evaluate the controlling of inbreeding in populations selected based on the best linear unbiased prediction (BLUP) considering a higher heritability value, it was simulated a genome with a heritability value higher than the real one. One base population of 1,000 individuals (500 males and 500 females) was simulated. From the base population, 20 males and 200 females were randomly selected and mated to produce 8 descendants per mating, producing 1,600 individuals. From the base population, the initial population was obtained, which generated the populations under selection. The selection was conducted over 50 generations, where the simulation process for each generation was repeated 20 times in order to minimize the effects of genetic drift (CARNEIRO, 2002). 20 males and 200 females were selected for each generation, and from the mating among them, 8 offspring were generated per mating. With the purpose to evaluate the effect of non-random mating in a population selected based on the best linear unbiased prediction (BLUP), populations with the following mating patterns were selected: Random mating (RM), exclusion of matings between full-sibs (FS) and half-sibs (HS), best to best mating and best to worst mating (compensatory mating). The mating patterns were compared by evaluating the following parameters across the generations: mean phenotypic value, average inbreeding, maximum inbreeding, losses due to fixation of unfavorable alleles, fixation of favorable alleles and selection limit. In populations selected through the best linear unbiased prediction (BLUP), the compensatory mating showed to be an effective tool for controlling inbreeding in the long run. The increase of the number of males in random matings, mating between the best breeders, which excludes the mating between HS and FS, and compensatory mating produced an increase of the phenotypic values in the long run due to the reduction of inbreeding, the selection limits and the rate of fixation of unfavorable alleles. Whether or not the females are selected considering the values found by BLUP does not change the need to control the inbreeding increments. Closed populations that do not select females also lose genetic diversity. The use of heritability higher than the real value was an effective method to reduce both the inbreeding increments, the fixation of unfavorable alleles and the loss of the selection limits. As a result of that, this method also produced higher phenotypic values.