Presents a Bayesian regression model of multiple level of complete genome and the Performance Prediction is compared with the popular model applied to each population separately Sembcorp aguas Santiago (clustering) and the set of Data Sets (clustering). For the small size of the population (e.g.< 50), the partial pooling increases the accuracy of the prediction of no clustering or grouping for the populations represented in the set of estimates. Clustering with partial models of multiple level can make optimal use of information in multiple sets of Population Estimates.Claim 1: a method for selecting individuals in a Breeding Program and for reproducing selected individuals The Method comprises: a) create a set of optimized Data for Prediction of complete genomes and, respectively,Select by: (i) Selecting and grouping sets of Predefined candidate Populations the predefined sets of Candidates selected breeding of crops, and selecting individuals of each Crossing of fenotipificaci\u00f3n cultivation for a comprehensive list of targeted individuals S for the selection.Where the genotypic information is available to all candidates (ii) to calculate, for a population of breeding of crops and traits of crops under artificial selection Active,Specific measures of Genetic Similarity or divergence with objective functions and estimated QTL effects of Population Specific features (iii) creating datasets of Best Estimate in terms of the number of Populations and individuals in each respective Population and I (V) use statistical models of no clustering.Partial or complete clustering clustering as a function of Genetic Similarity or divergence objective criteria (b) fenotipificar candidates in the Data Set of Best Estimate (c) to genotype individuals growing in a Plurality of markers (d) value S genomic estimated individuals genotyped for crop cultivation with the use of the fenotIPOs of the candidates in the datasets of Best Estimate (e) to make selections in the complete set
