Economical optimization of a breeding scheme by selective phenotyping of the calibration set in a multi-trait context: application to bread making quality.

KEY MESSAGE: Trait-assisted genomic prediction approach is a way to improve genetic gain by cost unit, by reducing budget allocated to phenotyping or by increasing the program's size for the same budget. This study compares different strategies of genomic prediction to optimize resource allocation in breeding schemes by using information from cheaper correlated traits to predict a more expensive trait of interest. We used bread wheat baking score (BMS) calculated for French registration as a case study. To conduct this project, 398 lines from a public breeding program were genotyped and phenotyped for BMS and correlated traits in 11 locations in France between 2000 and 2016. Single-trait (ST), multi-trait (MT) and trait-assisted (TA) strategies were compared in terms of predictive ability and cost. In MT and TA strategies, information from dough strength (W), a cheaper trait correlated with BMS (r = 0.45), was evaluated in the training population or in both the training and the validation sets, respectively. TA models allowed to reduce the budget allocated to phenotyping by up to 65% while maintaining the predictive ability of BMS. TA models also improved the predictive ability of BMS compared to ST models for a fixed budget (maximum gain: + 0.14 in cross-validation and + 0.21 in forward prediction). We also demonstrated that the budget can be further reduced by approximately one fourth while maintaining the same predictive ability by reducing the number of phenotypic records to estimate BMS adjusted means. In addition, we showed that the choice of the lines to be phenotyped can be optimized to minimize cost or maximize predictive ability. To do so, we extended the mean of the generalized coefficient of determination (CDmean) criterion to the multi-trait context (CDmulti).

Using crop growth model stress covariates and AMMI decomposition to better predict genotype-by-environment interactions.

KEY MESSAGE: We propose new methods to predict genotype × environment interaction by selecting relevant environmental covariates and using an AMMI decomposition of the interaction. Farmers are asked to produce more efficiently and to reduce their inputs in the context of climate change. They have to face more and more limiting factors that can combine in numerous stress scenarios. One solution to this challenge is to develop varieties adapted to specific environmental stress scenarios. For this, plant breeders can use genomic predictions coupled with environmental characterization to identify promising combinations of genes in relation to stress covariates. One way to do it is to take into account the genetic similarity between varieties and the similarity between environments within a mixed model framework. Molecular markers and environmental covariates (EC) can be used to estimate relevant covariance matrices. In the present study, based on a multi-environment trial of 220 European elite winter bread wheat (Triticum aestivum L.) varieties phenotyped in 42 environments, we compared reference regression models potentially including ECs, and proposed alternative models to increase prediction accuracy. We showed that selecting a subset of ECs, and estimating covariance matrices using an AMMI decomposition to benefit from the information brought by the phenotypic records of the training set are promising approaches to better predict genotype-by-environment interactions (G × E). We found that using a different kinship for the main genetic effect and the G × E effect increased prediction accuracy. Our study also demonstrates that integrative stress indexes simulated by crop growth models are more efficient to capture G × E than climatic covariates.

Predicting genomic selection efficiency to optimize calibration set and to assess prediction accuracy in highly structured populations.

KEY MESSAGE: We propose a criterion to predict genomic selection efficiency for structured populations. This criterion is useful to define optimal calibration set and to estimate prediction reliability for multiparental populations. Genomic selection refers to the use of genotypic information for predicting the performance of selection candidates. It has been shown that prediction accuracy depends on various parameters including the composition of the calibration set (CS). Assessing the level of accuracy of a given prediction scenario is of highest importance because it can be used to optimize CS sampling before collecting phenotypes, and once the breeding values are predicted it informs the breeders about the reliability of these predictions. Different criteria were proposed to optimize CS sampling in highly diverse panels, which can be useful to screen collections of genotypes. But plant breeders often work on structured material such as biparental or multiparental populations, for which these criteria are less adapted. We derived from the generalized coefficient of determination (CD) theory different criteria to optimize CS sampling and to assess the reliability associated to predictions in structured populations. These criteria were evaluated on two nested association mapping (NAM) populations and two highly diverse panels of maize. They were efficient to sample optimized CS in most situations. They could also estimate at least partly the reliability associated to predictions between NAM families, but they could not estimate differences in the reliability associated to the predictions of NAM families using the highly diverse panels as calibration sets. We illustrated that the CD criteria could be adapted to various prediction scenarios including inter and intra-family predictions, resulting in higher prediction accuracies.

Optimization of multi-environment trials for genomic selection based on crop models.

KEY MESSAGE: We propose a statistical criterion to optimize multi-environment trials to predict genotype × environment interactions more efficiently, by combining crop growth models and genomic selection models. Genotype × environment interactions (GEI) are common in plant multi-environment trials (METs). In this context, models developed for genomic selection (GS) that refers to the use of genome-wide information for predicting breeding values of selection candidates need to be adapted. One promising way to increase prediction accuracy in various environments is to combine ecophysiological and genetic modelling thanks to crop growth models (CGM) incorporating genetic parameters. The efficiency of this approach relies on the quality of the parameter estimates, which depends on the environments composing this MET used for calibration. The objective of this study was to determine a method to optimize the set of environments composing the MET for estimating genetic parameters in this context. A criterion called OptiMET was defined to this aim, and was evaluated on simulated and real data, with the example of wheat phenology. The MET defined with OptiMET allowed estimating the genetic parameters with lower error, leading to higher QTL detection power and higher prediction accuracies. MET defined with OptiMET was on average more efficient than random MET composed of twice as many environments, in terms of quality of the parameter estimates. OptiMET is thus a valuable tool to determine optimal experimental conditions to best exploit MET and the phenotyping tools that are currently developed.

Dent and Flint maize diversity panels reveal important genetic potential for increasing biomass production.

KEY MESSAGE: Genetic and phenotypic analysis of two complementary maize panels revealed an important variation for biomass yield. Flowering and biomass QTL were discovered by association mapping in both panels. The high whole plant biomass productivity of maize makes it a potential source of energy in animal feeding and biofuel production. The variability and the genetic determinism of traits related to biomass are poorly known. We analyzed two highly diverse panels of Dent and Flint lines representing complementary heterotic groups for Northern Europe. They were genotyped with the 50 k SNP-array and phenotyped as hybrids (crossed to a tester of the complementary pool) in a western European field trial network for traits related to flowering time, plant height, and biomass. The molecular information revealed to be a powerful tool for discovering different levels of structure and relatedness in both panels. This study revealed important variation and potential genetic progress for biomass production, even at constant precocity. Association mapping was run by combining genotypes and phenotypes in a mixed model with a random polygenic effect. This permitted the detection of significant associations, confirming height and flowering time quantitative trait loci (QTL) found in literature. Biomass yield QTL were detected in both panels but were unstable across the environments. Alternative kinship estimator only based on markers unlinked to the tested SNP increased the number of significant associations by around 40% with a satisfying control of the false positive rate. This study gave insights into the variability and the genetic architectures of biomass-related traits in Flint and Dent lines and suggests important potential of these two pools for breeding high biomass yielding hybrid varieties.

Maximizing the reliability of genomic selection by optimizing the calibration set of reference individuals: comparison of methods in two diverse groups of maize inbreds (Zea mays L.).

Genomic selection refers to the use of genotypic information for predicting breeding values of selection candidates. A prediction formula is calibrated with the genotypes and phenotypes of reference individuals constituting the calibration set. The size and the composition of this set are essential parameters affecting the prediction reliabilities. The objective of this study was to maximize reliabilities by optimizing the calibration set. Different criteria based on the diversity or on the prediction error variance (PEV) derived from the realized additive relationship matrix-best linear unbiased predictions model (RA-BLUP) were used to select the reference individuals. For the latter, we considered the mean of the PEV of the contrasts between each selection candidate and the mean of the population (PEVmean) and the mean of the expected reliabilities of the same contrasts (CDmean). These criteria were tested with phenotypic data collected on two diversity panels of maize (Zea mays L.) genotyped with a 50k SNPs array. In the two panels, samples chosen based on CDmean gave higher reliabilities than random samples for various calibration set sizes. CDmean also appeared superior to PEVmean, which can be explained by the fact that it takes into account the reduction of variance due to the relatedness between individuals. Selected samples were close to optimality for a wide range of trait heritabilities, which suggests that the strategy presented here can efficiently sample subsets in panels of inbred lines. A script to optimize reference samples based on CDmean is available on request.