RESUMO
The goal of this work was to compare the effect of the accuracy and residual variance in genome wide selection using marker selection as well as using the effect of the indirect selection, using simulated and real data. In simulated data was used one sample with 200 individuals with 1,000 molecular markers in F2 population. The real data was obtained in maize with F2 population with 441 individuals and genotyping with 261 SSR markers. There was 11 traits evaluated (ear length, ear width, row number, kernels per row, 100-kernel weight, ear weight, grain yield, length of branch, number of branch, plant height and ear height). All data was analyzed using rrBLUP method and 10-fold cross-validation. In simulated and maize data the results were similar: the residual variance with few markers is lower than with the 1000 markers and the accuracy with few markers is bigger than with 1000 markers. For maize data multi trait selection, the accuracy increased when the correlation between traits is greater than 0.50 and residual variance decreased when the correlation is greater than 0.70. In this sense, these results showed that marker selection could be used as a first step in genome wide selection, improving the prediction and compute demand.
O objetivo deste trabalho foi comparar o efeito da precisão e da variância residual na seleção genômica ampla utilizando a seleção de marcadores, bem como utilizando o efeito da seleção indireta, utilizando dados simulados e reais. Foram usados simulados de uma amostra com 200 indivíduos com 1.000marcadores moleculares na população F2. Os dados reais foram obtidos em milho com população F2 com 441indivíduos e genotipagem com 261 marcadores SSR. Foram avaliados 11 caracteres (comprimento da espiga,largura da espiga, número da linha, grãos por linha, peso de 100 grãos, peso da espiga, produtividade de grãos, comprimento da espiga, número de espigas, altura da planta e altura da espiga). Todos os dados foram analisados usando o método rrBLUP, sendo realizada 10 vezes a validação cruzada. Em dados simulados e de milho, os resultados foram semelhantes: a variância residual com poucos marcadores é menor do que com os marcadores 1000 e a precisão com poucos marcadores é maior do que com os marcadores 1000. Para a seleção multi-característica dos dados do milho, a precisão aumentou quando a correlação entre as características é maior que 0,50 e a variância residual diminuiu quando a correlação é maior que 0,70. Nesse sentido, esses resultados mostraram que a seleção de marcadores poderia ser usada como um primeiro passo na seleção genômica ampla, melhorando a previsão e a demanda computacional.
Assuntos
Zea mays , Melhoramento VegetalRESUMO
Population size and phenotypic measurement are two key factors determining the detection power of quantitative trait loci (QTL) mapping. We evaluated how these two controllable factors quantitatively affect the detection of QTL and their localization using a large F2 murine mapping population and found that three main points emerged from this study. One finding was that the sensitivity of QTL detection significantly decreased as the population size decreased. The decrease in the percentage logarithm of the odd score (LOD score, which is a statistical measure of the likelihood of two loci being lied near each other on a chromosome) can be estimated using the formula 1 - n/N, where n is the smaller and N the larger population size. This empirical formula has several practical implications in QTL mapping. We also found that a population size of 300 seems to be a threshold for the detection of QTL and their localization, which challenges the small population sizes commonly-used in published studies, in excess of 60 percent of which cite population sizes <300. In addition, it seems that the precision of phenotypic measurement has a limited capacity to affect detection power, which means that quantitative traits that cannot be measured precisely can also be used in QTL mapping for the detection of major QTL.