Eficiência de dialelos circulantes via simulação por reamostragem de um dialelo completo

Geralmente, o número de progenitores incluídos num dialelo completo é elevado, acarretando em muitas polinizações manuais, o que aliado às dificuldades das operações de campo, disponibilidade de recursos financeiros, mão-de-obra e período para a execução dos cruzamentos, pode tornar limitante a exec...

Nível de Acesso:openAccess
Publication Date:2003
Main Author: Ferreira, Fábio Medeiros lattes
Orientador/a: Ribeiro Júnior, José Ivo
Format: Dissertação
Language:por
Published: Universidade Federal de Viçosa
Áreas de Conhecimento:
Online Access:http://www.locus.ufv.br/handle/123456789/10462
Citação:FERREIRA, Fábio Medeiros. Eficiência de dialelos circulantes via simulação por reamostragem de um dialelo completo. 2003. 77 f. Dissertação (Mestrado em Genética e Melhoramento) - Universidade Federal de Viçosa, Viçosa. 2003.
Resumo inglês:In maize breeding programs there is almost always a huge number of progenitors to be included in a complete diallel, thus many manual pollinations need to be conducted. In addition, there are the usual difficulties of field operations, availability of financial resources, workmanship and the restricted period available to execute the crossings. Hence, the execution of a complete diallel may become unrealistic and incomplete diallels such as the circulant diallel seem to be a good alternative to the problems mentioned. In the present work we study estimation of the general combining ability (GCA) and specific combining ability (SCA) (GRIFFING, 1956) obtained from circulant diallels as compared to the analogous estimates obtained from the complete diallel. In a first study we simulated data for a complete diallel in order to study gene action in conection with circulant diallel effect. In a second study we used data from a complete diallel conducted by Pacheco (1997). In both studies the complete diallel involved 28 progenitors (p = 28) and their respective 378 F 1 hybrids (p(p- 1)/2). The circulant diallels were obtained from resampling of the complete diallel and each one of the 28 progenitors was represented by s crossings. The s values were set equal to 25, 23, 21, 19, 15, 13, 11, 9, 7, 5 and 3 and we generated 1000 circulant diallels for each one of these 12 alternative values of s. In the first study we simulated different degree of dominance to additive relations (d/a) of gene action using the values of d/a equal to 0, 0,5, 1,0, 1,5 and 2,0. In both studies we defined genetic parameters in order to evaluate the sample distribution of GCA e SCA estimated from the 12000 circulant diallels. We estimated Bootstrap expected values and sample standard deviations for all of the genetic parameters and considered the estimates obtained by using the complete diallel data (s = 27) as reference values (or parametric values). Simulation of circulant diallel data from resampling of the complete diallel data showed to be a very good alternative for comparing these schemes. Estimated values of GCA e SCA obtained from circulant diallels were similar to those obtained from the complete diallel except for a few progenitors in cases where a small s value was used. It was verified for the variable ear weight (t/ha), that with s = 19, equivalent to 30% reduction in the number of crosses, the circulant diallel allowed selection of the best progenitors and hybrid crosses in accordance with the complete diallel. As an overall conclusion we state that circulant diallel can be a good alternative to a complete diallel depending of the characteristics under evaluation, type of populations to be used and also on the goals of the breeding program.