Estratégias de seleção combinando informação individual e de família utilizando simulação de dados

• Ano de defesa: 2007
• Autor(a) principal: Santos, Lidiane Gomes dos
• Orientador(a): Euclydes, Ricardo Frederico
• Banca de defesa: Guimarães, Simone Eliza Facioni , Carneiro, Antônio Policarpo Souza
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Viçosa
• Programa de Pós-Graduação: Mestrado em Zootecnia
• Departamento: Genética e Melhoramento de Animais Domésticos; Nutrição e Alimentação Animal; Pastagens e Forragicul
• País: BR
• Palavras-chave em Português: Seleção combinada; Seleção individual; Seleção de famílias; Genesys; Análise de agrupamento; Método de Tocher
• Palavras-chave em Inglês: Combined selection; Individual selection; Family selection; Genesys; Cluster analysis; Tocher method
• Área do conhecimento CNPq: CNPQ::CIENCIAS AGRARIAS::ZOOTECNIA::GENETICA E MELHORAMENTO DOS ANIMAIS DOMESTICOS
• Link de acesso: http://locus.ufv.br/handle/123456789/5863

The superior phenotypes selection, in individual or families, is a practice of considerable importance for the improvers , once to obtain improved populations it is necessary to select and matings among or inside of families. This work s aim was to compare the efficiency of some arrangements between the individual selection and the selection among families to get combined selection indexes and soon afterwards to verify a cluster analysis was capable to detect different groups. Through the program GENESYS, it was simulated four genomes, each one considering only one characteristic, varying to each other just in value of heritability; then, we have two characteristics of low heritability (0,20 and 0,10), one of medium heritability (0,40) and one with high heritability (0,60). Using those genomes, bases populations of 1000 individuals (500 males and 500 females) with a rate endogamic equivalent to zero were simulated. In each base population, it was aleatory selected 10 males and 100 females corresponding, respectively, to 2% and 20% of selection intensity. When these individuals selected couple, it was maintained a control of 10 females for each male and, as a result, it was kept an initial population with 1000 descending, also maintaining a balance of 10 descending for female. For each characteristic selections, different arrangements were done to obtain combined selection s indexes. Thus, the first obtained index considers 100% for individual phenotype (Individual); the second considers 90% for individual phenotype and 10% for average phenotypic of the family (09P+01F); the third considers 70% for individual phenotype and 30% for average phenotypic of the family (07P+03F); the fourth considers 50% for the two values (05P+05F); the fifth considers 30% for individual phenotype and 70% for average phenotypic of the family (03P+07F); the sixth considers 10% for individual phenotype and 90% for average phenotypic of the family (01P+09F) and the seventh index considers 100% for the average phenotypic of the family (Family). The selected individuals' matings were driven at random. A final population of 1000 individuals was totaled. Each selection s strategy was led by 20 consecutive generations with 10 repetitions seeking to minimize the mistakes and effects of the genetic flotation. For a larger clarity in the interpretation of the results, a grouping analysis was accomplished by optimization by the method of Tocher in three intervals of time, with the first five generations, with the first ten generations and with the 20 total generations. The behavior of the selection s strategies for different herdability s values was not the same. As smaller the value of the heritability is, the difference among the averages phenotypics in each generation is bigger and the total earnings in these averages are smaller. The combination of the individual value with the family s average was more efficient for the lowest heritabilities. For these heritabilities, the individual selection or among families did not show to be a good option. The selection just through the individual s phenotypic value can be a good alternative for the characteristic of average herdabilidade and it was the best to the characteristic of high heritabilities.