The use of unbalanced historical data for genomic selection in an international wheat breeding program

Dawson, Julie; Endelman, Jeffrey B.; Heslot, Nicolas; Crossa, J.; Poland, Jesse; Dreisigacker, Susanne; Manès, Yann; Sorrells, Mark E.; Jannink, Jean‐Luc

doi:10.1016/j.fcr.2013.07.020

Cited by 99 publications

(118 citation statements)

References 31 publications

Supporting

Mentioning

104

Contrasting

Order By: Relevance

“…We compared two strategies to deal with GEI: reduce it (i.e., predict overall performance for a set of environments, GBLUP M ) and exploit it (i.e., predict performance by environment using the correlations among environments, GBLUP g´e ). We obtained high overall predictions accuracies for grain yield (GBLUP M ) compared with other studies (Heffner et al, 2011;Burgueño et al, 2012;Poland et al, 2012b;Charmet and Storlie, 2012;Dawson et al, 2013;Lado et al, 2013;Crossa et al, 2014;Lopez-Cruz et al, 2015). This probably is due to the large population size with extensive phenotyping and high quality experimental design with two to four replications and high heritability.…”

Section: Predictive Abilitymentioning

confidence: 56%

“…We found that genotype ´ year interactions contributed a large proportion of the GEI in our study. Year-to-year variations are more difficult to predict (Dawson et al, 2013) and usually contribute to a large proportion of the GEI (Ceretta and van Eeuwijk, 2008;Heslot et al, 2014). In our study, years 2010, 2011, and 2013 were high-yielding environments (i.e., yield average of 5370 kg ha ).…”

Section: Genotype ´ Environment Interaction Characterizationmentioning

confidence: 64%

“…One of the most challenging issues in plant breeding is the use of highly unbalanced existing phenotyping in the presence of GEI (Burgueño et al, 2012;Dawson et al, 2013;Heslot et al, 2014). Breeding programs produce large MET datasets that are highly unbalanced and with different levels of quality because of the degree of replication in different trials (Bernardo, 2010).…”

Section: Discussionmentioning

confidence: 99%

“…Most research on genomic selection has focused on accurately predicting the genotypic mean performance over all environments, assuming that all trial locations belong to the same target population of environments (Asoro et al, 2011;Charmet and Storlie, 2012;Storlie and Charmet, 2013;Crossa et al, 2013;Dawson et al, 2013). However, since GEI is widespread, the accuracy of genomic prediction could be lowered especially by the presence of crossover interaction (Burgueño et al, 2012;Dawson et al, 2013;Jarquín et al, 2014). Mixed models are powerful tools to make predictions by modeling the variancecovariance structure across environments to account for GEI (Piepho, 1997;Burgueño et al, 2008;Meyer, 2009;Malosetti et al, 2013).…”

Section: Predictive Abilitymentioning

confidence: 99%

“…Additionally, GS studies have included GEI information by performing overall predictions across environments (Heffner et al, 2011;Resende et al, 2011), within environments or groups of environments (Burgueño et al, 2012;Ly et al, 2013;Dawson et al, 2013;Heslot et al, 2014), or using marker-by-environment predictions (Jarquín et al, 2014;Lopez-Cruz et al, 2015). While mixed models have successfully been used to model GEI in GS, the best strategy to predict new genotypes or new environments is not clear.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Modeling Genotype × Environment Interaction for Genomic Selection with Unbalanced Data from a Wheat Breeding Program

et al. 2016

View full text Add to dashboard Cite

Genomic selection (GS) has successfully been used in plant breeding to improve selection efficiency and reduce breeding time and cost. However, there is not a clear strategy on how to incorporate genotype × environment interaction (GEI) to GS models. Increased prediction accuracy could be achieved using mixed models to exploit GEI by borrowing information from other environments. The objective of this work was to compare strategies to exploit GEI in GS using mixed models. Specifically, we compared strategies to predict new genotypes by borrowing information from other environments modeling the correlation matrix across environments and to design sets of environments aiming for low GEI to predict genomic performance in new environments. We evaluated 1477 advanced wheat (Triticum aestivum L.) lines for yield in 35 location–year combinations genotyped with genotyping‐by‐sequencing (GBS). Mixed models were used to obtain either overall or by‐environment predictions for different sets of environments. Overall accuracy was high (0.5). Borrowing information from relatives evaluated in multiple environments and modeling the correlation matrix across environments was the best strategy to predict new genotypes. On the other hand, the best strategy for predicting the performance of genotypes in new environments was either to predict across locations for single years or to predict within defined mega‐environments (MEs) for any year or location. In summary, higher predictive ability was obtained by characterizing and by modeling GEI in the GS context.

show abstract