Different population sizes were simulated to estimate the phenotypic values in the selection assisted by markers for quantitative characteristics with heritability values of 0.10, 0.40 and 0.70. Cluster analysis with the phenotypic performance was carried out aiming at obtaining classification structure among the samples to optimize QTL detection. The genetic simulation system (Genesys) was used for the simulation of three genomes (each one consisting of a single characteristic whose distinction was the value of heritability), and the base and original populations. Each initial population was submitted to selection assisted by markers for 20 consecutive generations, in which selected parents mated selectively among the best and the worst ones. This selective strategy of mating proved itself to be efficient in reducing the number of individuals required in a population for QTL mapping. As the magnitude of heritability increases, lower population sizes are required to maintain similarities in phenotypic increments. The use of samples with 300, 250 and 200 individuals or more for heritabilities of 0.10, 0.40 and 0.70, respectively, is unnecessary, because of the equivalent inferences indicated by Tocher optmization and complete linkage methods from the system of statistical and genetic analysis (SAEG).
cluster analysis; heritability; sampling; simulation; QTL
