In the era of genome-wide selection (GWS), genotype-by-environment (G×E) interactions can be studied using genomic information, thus enabling the estimation of SNP marker effects and the prediction of genomic estimated breeding values (GEBV) for young candidates for selection in different environments. Although G×E studies in pigs are scarce, the use of artificial insemination has enabled the distribution of genetic material from sires across multiple environments. Given the relevance of reproductive traits, such as the total number born (TNB) and the variation in environmental conditions encountered by commercial dams, understanding G×E interactions can be essential for choosing the best sires for different environments. The present work proposes a two-step reaction norm approach for G×E analysis using genomic information. The first step provided estimates of environmental effects (herd-year-season, HYS), and the second step provided estimates of the intercept and slope for the TNB across different HYS levels, obtained from the first step, using a random regression model. In both steps, pedigree ( A: ) and genomic ( G: ) relationship matrices were considered. The genetic parameters (variance components, h(2) and genetic correlations) were very similar when estimated using the A: and G: relationship matrices. The reaction norm graphs showed considerable differences in environmental sensitivity between sires, indicating a reranking of sires in terms of genetic merit across the HYS levels. Based on the G: matrix analysis, SNP by environment interactions were observed. For some SNP, the effects increased at increasing HYS levels, while for others, the effects decreased at increasing HYS levels or showed no changes between HYS levels. Cross-validation analysis demonstrated better performance of the genomic approach with respect to traditional pedigrees for both the G×E and standard models. The genomic reaction norm model resulted in an accuracy of GEBV for "juvenile" boars varying from 0.14 to 0.44 across different HYS levels, while the accuracy of the standard genomic prediction model, without reaction norms, varied from 0.09 to 0.28. These results show that it is important and feasible to consider G×E interactions in evaluations of sires using genomic prediction models and that genomic information can increase the accuracy of selection across environments.

BackgroundA complete approach for genome-wide selection (GWS) involves reliable statistical genetics models and methods. Reports on this topic are common for additive genetic models but not for additive-dominance models. The objective of this paper was (i) to compare the performance of 10 additive-dominance predictive models (including current models and proposed modifications), fitted using Bayesian, Lasso and Ridge regression approaches; and (ii) to decompose genomic heritability and accuracy in terms of three quantitative genetic information sources, namely, linkage disequilibrium (LD), co-segregation (CS) and pedigree relationships or family structure (PR). The simulation study considered two broad sense heritability levels (0.30 and 0.50, associated with narrow sense heritabilities of 0.20 and 0.35, respectively) and two genetic architectures for traits (the first consisting of small gene effects and the second consisting of a mixed inheritance model with five major genes).ResultsG-REML/G-BLUP and a modified Bayesian/Lasso (called BayesA*B* or t-BLASSO) method performed best in the prediction of genomic breeding as well as the total genotypic values of individuals in all four scenarios (two heritabilities x two genetic architectures). The BayesA*B*-type method showed a better ability to recover the dominance variance/additive variance ratio. Decomposition of genomic heritability and accuracy revealed the following descending importance order of information: LD, CS and PR not captured by markers, the last two being very close.ConclusionsAmongst the 10 models/methods evaluated, the G-BLUP, BAYESA*B* (−2,8) and BAYESA*B* (4,6) methods presented the best results and were found to be adequate for accurately predicting genomic breeding and total genotypic values as well as for estimating additive and dominance in additive-dominance genomic models.

No abstract

RESUMOAvaliou-se com este trabalho o comportamento de 28 progênies F3 de Coffea arabica obtidas de cruzamentos das cultivares Catuaí Vermelho e Catuaí Amarelo com descendentes do Híbrido de Timor, realizados pela Epamig em conjunto com a UFV. Estimaram-se os parâmetros genéticos e as correlações entre caracteres, buscando conhecer a estrutura genética da população e seu potencial para o melhoramento. O experimento foi instalado em Patrocínio, Estado de Minas Gerais, em delineamento de blocos casualizados com seis repetições, contendo 28 progênies F3 e duas testemunhas da cultivar Catuaí Vermelho IAC 15. Analisaram-se as produções obtidas nas quatro colheitas iniciais, de 1997 a 2000 e alguns caracteres vegetativos. As progênies avaliadas apresentaram média de produção de grãos superior à testemunha, e grande variabilidade genética, sugerindo a possibilidade de se obter linhagens superiores. As progênies avaliadas apresentaram-se resistentes às raças fisiológicas da ferrugem presentes na região do experimento. A progênie 505-9-2 se destaca como material produtivo e vigoroso e de porte alto, enquanto as progênies 514-7-10 e 514-7-6 além de produtivas e vigorosas apresentaram porte baixo semelhante à cultivar Catuaí Vermelho IAC 15. A produção de grãos apresentou correlação genotípica alta e positiva com os caracteres diâmetro do caule, vigor, porte, altura e diâmetro da planta, mas não apresentou resultados consistentes de correlação com carga pendente. A produção de grãos de anos de colheita mostrou-se correlacionada com a produção total de quatro anos apenas a partir do segundo ano de produção.Palavras-chave: café, Coffea arabica L., melhoramento do cafeeiro, Híbrido de Timor.

It was studied the parametric restrictions of the diallel analysis model of Griffing, method 2 (parents and F 1 generations) and model 1 (fixed), in order to address the questions: i) does the statistical model need to be restricted? ii) do the restrictions satisfy the genetic parameter values? and iii) do they make the analysis and interpretation easier? Objectively, these questions can be answered as: i) yes, ii) not all of them, and iii) the analysis is easier, but the interpretation is the same as in the model with restrictions that satisfy the parameter values. The main conclusions were that: the statistical models for combining ability analysis are necessarily restricted; in the Griffing model