Método de Haseman-Elston e suas modificações no estudo de genes controladores de característica quantitativa

• Autor(a) principal: Tomaz, Rafael Simões
• Data de Publicação: 2012
• Tipo de documento: Tese
• Idioma: por
• Título da fonte: LOCUS Repositório Institucional da UFV
• Texto Completo: http://locus.ufv.br/handle/123456789/1343
• topic: QTL; Haseman-Elston; Simulação; QTL; Haseman-Elston; Simulation; CNPQ::CIENCIAS BIOLOGICAS::GENETICA::GENETICA QUANTITATIVA

QTL mapping allows to study loci that play a role in complex traits, as well to determine their chromosomal locations. For this purpose, a set of methods useful for the study of quantitative traits are those based on Haseman-EIston regression, that is quite used in human genetics. These are robustness regression methods for linkage analysis that can be used in order to associate genetic markers with phenotypic trait. With SNP genetic markers, that can be genotyped within the coding regions, such methods had became an alternative to detect QTLs. Such methods have a potential as they constitute an efficient tool to of evaluating the genetic data from exogamic crosses. However, It is necessary a better understanding of the properties of such methodologies, in terms of the detection process as well as its power. Thus, this study proposes the investigation of sib-pair analysis based on Haseman-Elston regression, by means of simulated data. We evaluated the original Haseman-Elston regression (oHE), in two selective schemes, and revisited Haseman-Elston regression (rHE) in order to provide theoretical basis for using exogamic populations for studies of gene identification. Data simulation procedure was performed with GQMOL software. It was simulated QTLs with different gene actions (1, 3, 5, 7, 10, 13, 16, 20, 25, 27, 30 and 43%) and traits with different heritabilities (35, 50, 65 and 80%) under different size population (25, 50, 75, 100 and 200 sibs) in order to evaluate the statistical power of methods. Data analysis was performed using the free software R for Statistical Computing. All methods have proved ineffective in the study of quantitative traits, but had a great potential for the study of oligogenic traits. The method rHE showed high potential to detect major effect genes in populations larger than a hundred sibs in traits with heritability greater than 50%. However, all methods showed themselves ineffective to detect low effect genes.