ABSTRACT. This work aimed to present the optimum environment number methodology and propose the optimization of the National Sunflower Trials Network, by means of the environments exclusion that do not provide loss of the environmental variability already established. Grain and oil yield data of 16 genotypes evaluated at 16 environments of the National Sunflower Trials Network, obtained from trials conducted out-of-season in 2012 and 2013 were used. An analysis was proposed to establish the optimum environment number for genotypes evaluation, based on genotype performance in the various environmental combinations. The removal or maintenance of environments in the experimental network was dynamic, since different environmental combinations impacted the representativeness of the complete network in a different way. This analysis also provides a graphical view of the impact of the environment removal from the network. Once detected points below the established correlation, the researcher could infer about the network minimum environment number and, suggest through consistent information of several testing years, the environment exclusion.

RESUMO Visando auxiliar o melhorista na escolha da melhor metodologia para análises de diversidade genética em populações de retrocruzamento avaliaram-se os métodos baseados em informações fenotípicas. Os fenótipos foram simulados para 13 características, geradas em 10 populações com 100 indivíduos cada. Geraram-se informações genotípicas de 100 locos, dos quais 20 foram tomados ao acaso para determinar as características, manifestando dois alelos. Medidas de dissimilaridade foram calculadas, e analisou-se a diversidade genética por meio agrupamento hierárquico e projeção gráfica das distâncias. A partir das duas populações mais divergentes fez-se o retrocruzamento. Considerou-se um conjunto de características com herdabilidade variável. Simulou-se o efeito ambiental admitindo distribuição normal, com média zero e variância σ2. Para as análises de agrupamentos hierárquicos utilizaram-se os métodos: Método de Gower, Ligação média dentro de grupo, Ligação média entre grupo, Vizinho mais distante, Vizinho mais próximo, Método de Ward, Método da mediana. O efeito ambiental e a herdabilidade das variáveis analisadas influenciaram o padrão de agrupamento de populações sob retrocruzamento. Em características de herdabilidade elevada, o método do Vizinho mais próximo foi o mais eficiente em reconstituir o retrocruzamento, além de ter apresentado a maior correlação cofenética, sendo considerada a melhor metodologia a priori e a posteriori.

ABSTRACT Methods to obtain phenotypic information were evaluated to help breeders choosing the best methodology for analysis of genetic diversity in backcross populations. Phenotypes were simulated for 13 characteristics generated in 10 populations with 100 individuals each. Genotypic information was generated from 100 loci of which 20 were taken at random to determine the characteristics expressing two alleles. Dissimilarity measures were calculated, and genetic diversity was analyzed through hierarchical clustering and graphic projection of the distances. A backcross was performed from the two most divergent populations. A set of characteristics with variable heritability was taken into account. The environmental effect was simulated assuming x~N(0, σ2). For hierarchical clusters, the following methods were used: Gower Method, average linkage within the cluster, average linkage among clusters, the furthest neighbor method, the nearest neighbor method, Ward's method, and the median method. The environmental effect and heritability of the analyzed variables had an influence on the pattern of hierarchical clustering populations according to the backcrossed generations. The nearest neighbor method was the most efficient in reconstructing the system of backcrossing, and it presented the highest cophenetic correlation. The efficiency of the nearest neighbor method was the highest when the analysis involved characteristics of high heritability.