The objective of this study was to estimate the results ofthe upland rice {Oryza sativa L.) breeding program conducted by the Brazilian Agricultural Research Corporation (Embrapa) and collaborators over the period of 1984 to 2009 covering 25 annual steps of improvement. The best lines generated by this program are evaluated in "value for cultivation and use (VCU) trials." This study used data from 603 VCU trials conducted in seven Brazilian States. The group of lines entering VCU in each year was faken as a sample of the elife program in that year. Best linear unbiased estimates (BLUEs) of the means of groups were computed, and the regression of the BLUEs on years was taken as an estimate of the efficiency of the breeding program.

No abstract

Os ganhos genéticos para produtividade obtidos pelo programa de melhoramento do arroz irrigado por inundação na Região Nordeste do Brasil no período de 1984 a 1993 foram estimados visando avaliar a eficiência do programa e traçar novas estratégias. Esta estimativa foi feita a partir dos dados de 59 ensaios regionais de rendimento conduzidos pelas empresas de pesquisa agropecuária do Nordeste, em cooperação com a Embrapa-Centro Nacional de Pesquisa de Arroz e Feijão (CNPAF), Goiânia, GO. O método estatístico utilizado baseia-se em médias ajustadas por modelo linear generalizado. O ganho genético médio estimado foi de 54,9 ± 14,4 kg/ha/ano (0,8%). Nos últimos três anos houve uma tendência de interrupção dos ganhos. A pequena magnitude dos ganhos para produção nesta região podem ser atribuídos ao direcionamento do programa de geração de linhagens da Embrapa para qualidade de grãos e resistência a doenças, às diferenças ambientais existentes entre Goiânia e a Região Nordeste e ao pequeno número de ensaios conduzidos. A genealogia das linhagens foi traçada e verificou-se que os principais ancestrais são os mesmos das cultivares recomendadas. A base genética das linhagens é estreita, o que também pode estar contribuindo para a obtenção de pequenos ganhos genéticos para produtividade.

The relative performance of one genotype is not identical in different environments due to genotype-environment interaction (G9E). Thus, for a breeding program to successfully develop cultivars, it is fundamental that candidate elite-lines are tested in several target environments and that the data are analysed for yield, adaptability and stability. The objective of this work was to study the G9E for upland rice using a mixed model and, using the harmonic mean of relative performance of genotypic values (HMRPGV) method, to analyse cultivars and elite-lines over time to identify those that aggregate high grain yield (GY) with high genotypic adaptability and stability. A large dataset of ''value for cultivation and use trials'' collected by the Brazilian Agricultural Research Corporation (Embrapa) and collaborators from 1984 to 2010, involving seven states that represent upland rice crops in the Midwest, North and Northeast regions of Brazil, was used. The effect of location was shown to be more important than the effect of year for promoting crossover interaction. The CNA 8555 had the best GY associated with adaptability and stability, presenting a superiority of 13.28 % above the general mean of all elite-lines. Using already-released cultivars and potential elite-lines, the generalised linear regression analysis revealed significant progress of the stability and adaptability associated with GY over time. The HMRPGV method was shown to be an important tool and allowed identification of three elite-lines in the Embrapa pipeline (AB 062008, AB 062041 and AB 062037), each with high stability, adaptability and yield potential to be released commercially.Keywords Oryza sativa Á HMRPGV Á BLUP Á REML Á G9E Á Genetic progress Abbreviations BLUPBest linear unbiased predictor REML Restricted maximum likelihood GY Grain yield VCU Value for cultivation and use G9EGenotype-environment interaction G9LGenotype-location interaction G9YGenotype-year interaction G9L9YGenotype-location-year interaction HMRPGV Harmonic mean of relative performance of genotypic values

In recurrent selection programs, progeny testing is done in multienvironment trials, which generates genotype × environment interaction (G × E). Therefore, modeling G × E is essential for genomic prediction in the context of recurrent genomic selection (RGS). Developing single‐step, best linear unbiased prediction‐based reaction norm models (termed RN‐HBLUP) using data from nongenotyped and genotyped progenies, can enhance predictive accuracy. Our objectives were to evaluate: (i) a class of RN‐HBLUP models accommodating combined relationship of pedigree and genomic data, environmental covariates, and their interactions for prediction of phenotypic responses; (ii) the predictive accuracy of these models and the relative importance of main effects and interaction components; and (iii) the influence of different grouping strategies of genetic–environmental data (within selection cycles or across cycles) on prediction accuracy of the merit for untested progenies. The genetic material comprised 667 S1:3 progenies of irrigated rice (Oryza sativa L.) and six check cultivars. These materials were evaluated in yield trials conducted in 10 environments during three selection cycles. Genomic information was derived from single‐nucleotide polymorphism markers genotyped on 174 progenies in the third cycle. We evaluated six predictive models. Environmental covariates and G × E interaction explained a significant portion of the phenotypic variance, increasing accuracy and decreasing the bias of phenotypic prediction. Within‐cycle data were sufficient for accurate prediction of untested progenies, even in untested environments. We concluded that the RN‐HBLUP model, with the comprehensive structure, could be useful in improving the prediction accuracy of quantitative traits in RGS programs.