Enhanced radiation use efficiency and grain filling rate as the main drivers of grain yield genetic gains in the CIMMYT elite spring wheat yield trial.

Common wheat (Triticum aestivum L.) is a major staple food crop, providing a fifth of food calories and proteins to the world's human population. Despite the impressive growth in global wheat production in recent decades, further increases in grain yield are required to meet future demands. Here we estimated genetic gain and genotype stability for grain yield (GY) and determined the trait associations that contributed uniquely or in combination to increased GY, through a retrospective analysis of top-performing genotypes selected from the elite spring wheat yield trial (ESWYT) evaluated internationally during a 14-year period (2003 to 2016). Fifty-six ESWYT genotypes and four checks were sown under optimally irrigated conditions in three phenotyping trials during three consecutive growing seasons (2018-2019 to 2020-2021) at Norman E. Borlaug Research Station, Ciudad Obregon, Mexico. The mean GY rose from 6.75 (24th ESWYT) to 7.87 t ha-1 (37th ESWYT), representing a cumulative increase of 1.12 t ha-1. The annual genetic gain for GY was estimated at 0.96% (65 kg ha-1 year-1) accompanied by a positive trend in genotype stability over time. The GY progress was mainly associated with increases in biomass (BM), grain filling rate (GFR), total radiation use efficiency (RUE_total), grain weight per spike (GWS), and reduction in days to heading (DTH), which together explained 95.5% of the GY variation. Regression lines over the years showed significant increases of 0.015 kg m-2 year-1 (p < 0.01), 0.074 g m-2 year-1 (p < 0.05), and 0.017 g MJ-1 year-1 (p < 0.001) for BM, GFR, and RUE_total, respectively. Grain weight per spike exhibited a positive but no significant trend (0.014 g year-1, p = 0.07), whereas a negative tendency for DTH was observed (- 0.43 days year-1, p < 0.001). Analysis of the top ten highest-yielding genotypes revealed differential GY-associated trait contributions, demonstrating that improved GY can be attained through different mechanisms and indicating that no single trait criterion is adopted by CIMMYT breeders for developing new superior lines. We conclude that CIMMYT's Bread Wheat Breeding Program has continued to deliver adapted and more productive wheat genotypes to National partners worldwide, mainly driven by enhancing RUE_total and GFR and that future yield increases could be achieved by intercrossing genetically diverse top performer genotypes.

Genomic Prediction from Multi-Environment Trials of Wheat Breeding.

Genomic prediction relates a set of markers to variability in observed phenotypes of cultivars and allows for the prediction of phenotypes or breeding values of genotypes on unobserved individuals. Most genomic prediction approaches predict breeding values based solely on additive effects. However, the economic value of wheat lines is not only influenced by their additive component but also encompasses a non-additive part (e.g., additive × additive epistasis interaction). In this study, genomic prediction models were implemented in three target populations of environments (TPE) in South Asia. Four models that incorporate genotype × environment interaction (G × E) and genotype × genotype (GG) were tested: Factor Analytic (FA), FA with genomic relationship matrix (FA + G), FA with epistatic relationship matrix (FA + GG), and FA with both genomic and epistatic relationship matrices (FA + G + GG). Results show that the FA + G and FA + G + GG models displayed the best and a similar performance across all tests, leading us to infer that the FA + G model effectively captures certain epistatic effects. The wheat lines tested in sites in different TPE were predicted with different precisions depending on the cross-validation employed. In general, the best prediction accuracy was obtained when some lines were observed in some sites of particular TPEs and the worse genomic prediction was observed when wheat lines were never observed in any site of one TPE.

Data Augmentation Enhances Plant-Genomic-Enabled Predictions.

Genomic selection (GS) is revolutionizing plant breeding. However, its practical implementation is still challenging, since there are many factors that affect its accuracy. For this reason, this research explores data augmentation with the goal of improving its accuracy. Deep neural networks with data augmentation (DA) generate synthetic data from the original training set to increase the training set and to improve the prediction performance of any statistical or machine learning algorithm. There is much empirical evidence of their success in many computer vision applications. Due to this, DA was explored in the context of GS using 14 real datasets. We found empirical evidence that DA is a powerful tool to improve the prediction accuracy, since we improved the prediction accuracy of the top lines in the 14 datasets under study. On average, across datasets and traits, the gain in prediction performance of the DA approach regarding the Conventional method in the top 20% of lines in the testing set was 108.4% in terms of the NRMSE and 107.4% in terms of the MAAPE, but a worse performance was observed on the whole testing set. We encourage more empirical evaluations to support our findings.

Deep learning methods improve genomic prediction of wheat breeding.

In the field of plant breeding, various machine learning models have been developed and studied to evaluate the genomic prediction (GP) accuracy of unseen phenotypes. Deep learning has shown promise. However, most studies on deep learning in plant breeding have been limited to small datasets, and only a few have explored its application in moderate-sized datasets. In this study, we aimed to address this limitation by utilizing a moderately large dataset. We examined the performance of a deep learning (DL) model and compared it with the widely used and powerful best linear unbiased prediction (GBLUP) model. The goal was to assess the GP accuracy in the context of a five-fold cross-validation strategy and when predicting complete environments using the DL model. The results revealed the DL model outperformed the GBLUP model in terms of GP accuracy for two out of the five included traits in the five-fold cross-validation strategy, with similar results in the other traits. This indicates the superiority of the DL model in predicting these specific traits. Furthermore, when predicting complete environments using the leave-one-environment-out (LOEO) approach, the DL model demonstrated competitive performance. It is worth noting that the DL model employed in this study extends a previously proposed multi-modal DL model, which had been primarily applied to image data but with small datasets. By utilizing a moderately large dataset, we were able to evaluate the performance and potential of the DL model in a context with more information and challenging scenario in plant breeding.

A marker weighting approach for enhancing within-family accuracy in genomic prediction.

Genomic selection is revolutionizing plant breeding. However, its practical implementation is still very challenging, since predicted values do not necessarily have high correspondence to the observed phenotypic values. When the goal is to predict within-family, it is not always possible to obtain reasonable accuracies, which is of paramount importance to improve the selection process. For this reason, in this research, we propose the Adversaria-Boruta (AB) method, which combines the virtues of the adversarial validation (AV) method and the Boruta feature selection method. The AB method operates primarily by minimizing the disparity between training and testing distributions. This is accomplished by reducing the weight assigned to markers that display the most significant differences between the training and testing sets. Therefore, the AB method built a weighted genomic relationship matrix that is implemented with the genomic best linear unbiased predictor (GBLUP) model. The proposed AB method is compared using 12 real data sets with the GBLUP model that uses a nonweighted genomic relationship matrix. Our results show that the proposed AB method outperforms the GBLUP by 8.6, 19.7, and 9.8% in terms of Pearson's correlation, mean square error, and normalized root mean square error, respectively. Our results support that the proposed AB method is a useful tool to improve the prediction accuracy of a complete family, however, we encourage other investigators to evaluate the AB method to increase the empirical evidence of its potential.

Multivariate Genomic Hybrid Prediction with Kernels and Parental Information.

Genomic selection (GS) plays a pivotal role in hybrid prediction. It can enhance the selection of parental lines, accurately predict hybrid performance, and harness hybrid vigor. Likewise, it can optimize breeding strategies by reducing field trial requirements, expediting hybrid development, facilitating targeted trait improvement, and enhancing adaptability to diverse environments. Leveraging genomic information empowers breeders to make informed decisions and significantly improve the efficiency and success rate of hybrid breeding programs. In order to improve the genomic ability performance, we explored the incorporation of parental phenotypic information as covariates under a multi-trait framework. Approach 1, referred to as Pmean, directly utilized parental phenotypic information without any preprocessing. While approach 2, denoted as BV, replaced the direct use of phenotypic values of both parents with their respective breeding values. While an improvement in prediction performance was observed in both approaches, with a minimum 4.24% reduction in the normalized root mean square error (NRMSE), the direct incorporation of parental phenotypic information in the Pmean approach slightly outperformed the BV approach. We also compared these two approaches using linear and nonlinear kernels, but no relevant gain was observed. Finally, our results increase empirical evidence confirming that the integration of parental phenotypic information helps increase the prediction performance of hybrids.

Do feature selection methods for selecting environmental covariables enhance genomic prediction accuracy?

Genomic selection (GS) is transforming plant and animal breeding, but its practical implementation for complex traits and multi-environmental trials remains challenging. To address this issue, this study investigates the integration of environmental information with genotypic information in GS. The study proposes the use of two feature selection methods (Pearson's correlation and Boruta) for the integration of environmental information. Results indicate that the simple incorporation of environmental covariates may increase or decrease prediction accuracy depending on the case. However, optimal incorporation of environmental covariates using feature selection significantly improves prediction accuracy in four out of six datasets between 14.25% and 218.71% under a leave one environment out cross validation scenario in terms of Normalized Root Mean Squared Error, but not relevant gain was observed in terms of Pearson´s correlation. In two datasets where environmental covariates are unrelated to the response variable, feature selection is unable to enhance prediction accuracy. Therefore, the study provides empirical evidence supporting the use of feature selection to improve the prediction power of GS.

Optimizing Sparse Testing for Genomic Prediction of Plant Breeding Crops.

While sparse testing methods have been proposed by researchers to improve the efficiency of genomic selection (GS) in breeding programs, there are several factors that can hinder this. In this research, we evaluated four methods (M1-M4) for sparse testing allocation of lines to environments under multi-environmental trails for genomic prediction of unobserved lines. The sparse testing methods described in this study are applied in a two-stage analysis to build the genomic training and testing sets in a strategy that allows each location or environment to evaluate only a subset of all genotypes rather than all of them. To ensure a valid implementation, the sparse testing methods presented here require BLUEs (or BLUPs) of the lines to be computed at the first stage using an appropriate experimental design and statistical analyses in each location (or environment). The evaluation of the four cultivar allocation methods to environments of the second stage was done with four data sets (two large and two small) under a multi-trait and uni-trait framework. We found that the multi-trait model produced better genomic prediction (GP) accuracy than the uni-trait model and that methods M3 and M4 were slightly better than methods M1 and M2 for the allocation of lines to environments. Some of the most important findings, however, were that even under a scenario where we used a training-testing relation of 15-85%, the prediction accuracy of the four methods barely decreased. This indicates that genomic sparse testing methods for data sets under these scenarios can save considerable operational and financial resources with only a small loss in precision, which can be shown in our cost-benefit analysis.

Grain yield genetic gains and changes in physiological related traits for CIMMYT's High Rainfall Wheat Screening Nursery tested across international environments.

The effects of climate change together with the projected future demand represents a huge challenge for wheat production systems worldwide. Wheat breeding can contribute to global food security through the creation of genotypes exhibiting stress tolerance and higher yield potential. The objectives of our study were to (i) estimate the annual grain yield (GY) genetic gain of High Rainfall Wheat Yield Trials (HRWYT) grown from 2007 (15th HRWYT) to 2016 (24th HRWYT) across international environments, and (ii) determine the changes in physiological traits associated with GY genetic improvement. The GY genetic gains were estimated as genetic progress per se (GYP) and in terms of local checks (GYLC). In total, 239 international locations were classified into two groups: high- and low-rainfall environments based on climate variables and trial management practices. In the high-rainfall environment, the annual genetic gains for GYP and GYLC were 3.8 and 1.17 % (160 and 65.1 kg ha-1 yr-1), respectively. In the low-rainfall environment, the genetic gains were 0.93 and 0.73 % (40 and 33.1 kg ha-1 yr-1), for GYP and GYLC respectively. The GY of the lines included in each nursery showed a significant phenotypic correlation between high- and low-rainfall environments in all the examined years and several of the five best performing lines were common in both environments. The GY progress was mainly associated with increased grain weight (R2 = 0.35 p < 0.001), days to maturity (R2 = 0.20, p < 0.001) and grain filling period (R2 = 0.06, p < 0.05). These results indicate continuous GY genetic progress and yield stability in the HRWYT germplasm developed and distributed by CIMMYT.