Prospects for increasing yield in macadamia using component traits and genomics

Selection of candidate cultivars in macadamia requires extensive phenotypic measurements over many years and trials. In particular, yield traits such as nut-in-shell yield and kernel yield are economically vital characteristics and therefore guide the selection process for new cultivars. However, these traits can only be measured in mature trees, resulting in long generation intervals and slow rates of genetic gain. In addition, these traits are expensive to measure. Strategies to reduce the generation interval and increase the intensity of selection include using yield component traits, identification of markers associated with component traits, and genomic selection for yield. Yield component traits that contribute to resource availability for fruit formation include floral and nut characteristics. In this review, these traits will be investigated to estimate their relative importance in macadamia breeding and their heritability and correlations with yield. Furthermore, the usefulness of genome-wide association studies regarding yield component traits will be reviewed. Genetic-based breeding techniques could exploit this information to increase yield gains per breeding cycle and estimate the quantitative nature of yield traits. Genomic selection uses genome-wide molecular markers to predict the phenotype of individuals at an early age before maturity, thereby reducing the cycle time and increasing gain per unit time in plant breeding programmes. This review evaluates the potential for measurement of yield component traits, genome-wide association studies, and genomic selection to be employed in the Australian macadamia breeding programme to accelerate gains for nut yield.     

Genetic analysis for physical nut traits in almond

Almond breeding is increasingly taking into account kernel quality as a breeding objective. Although information on nut and kernel physical parameters involved in almond quality has already been compiled, the genetic control of these traits has not been studied. This genetic information would improve the efficacy of almond breeding programs. A linkage map with 56 simple-sequence repeat markers was constructed for the “Vivot” × “Blanquerna” almond population showing a wide range of variability for the physical parameters of nut and kernel. A total of 14 putative quantitative trait loci (QTLs) controlling these physical traits were detected in the current study, corresponding to six genomic regions of the eight almond linkage groups (LG). Some QTLs are colocated in the same region or shared the same molecular markers, in a manner that reflects the correlations between the physical traits, as well as with the chemical components of the almond kernel. The limit of detection values for any given trait ranged from 2.06 to 5.17, explaining between 13.0 and 44.0 % of the phenotypic variance of the trait. This new genetic information needs to be taken into account when breeding for physical traits in almond. Increases in the positive quality traits, both physical and chemical, need to be considered simultaneously whenever they are genetically independent, even if they are negatively correlated. This is the first complete genetic framework map for physical components of almond nut and kernel, with 14 putative QTLs associated with a large number of parameters controlling physical traits in almond.     

Accelerating the Switchgrass (Panicum virgatum L.) Breeding Cycle Using Genomic Selection Approaches

Switchgrass (Panicum virgatum L.) is a perennial grass undergoing development as a biofuel feedstock. One of the most important factors hindering breeding efforts in this species is the need for accurate measurement of biomass yield on a per-hectare basis. Genomic selection on simple-to-measure traits that approximate biomass yield has the potential to significantly speed up the breeding cycle. Recent advances in switchgrass genomic and phenotypic resources are now making it possible to evaluate the potential of genomic selection of such traits. We leveraged these resources to study the ability of three widely-used genomic selection models to predict phenotypic values of morphological and biomass quality traits in an association panel consisting of predominantly northern adapted upland germplasm. High prediction accuracies were obtained for most of the traits, with standability having the highest ten-fold cross validation prediction accuracy (0.52). Moreover, the morphological traits generally had higher prediction accuracies than the biomass quality traits. Nevertheless, our results suggest that the quality of current genomic and phenotypic resources available for switchgrass is sufficiently high for genomic selection to significantly impact breeding efforts for biomass yield.     

Accuracy of genomic selection methods in a standard data set of loblolly pine (Pinus taeda L.)

Genomic selection can increase genetic gain per generation through early selection. Genomic selection is expected to be particularly valuable for traits that are costly to phenotype and expressed late in the life cycle of long-lived species. Alternative approaches to genomic selection prediction models may perform differently for traits with distinct genetic properties. Here the performance of four different original methods of genomic selection that differ with respect to assumptions regarding distribution of marker effects, including (i) ridge regression-best linear unbiased prediction (RR-BLUP), (ii) Bayes A, (iii) Bayes Cπ, and (iv) Bayesian LASSO are presented. In addition, a modified RR-BLUP (RR-BLUP B) that utilizes a selected subset of markers was evaluated. The accuracy of these methods was compared across 17 traits with distinct heritabilities and genetic architectures, including growth, development, and disease-resistance properties, measured in a Pinus taeda (loblolly pine) training population of 951 individuals genotyped with 4853 SNPs. The predictive ability of the methods was evaluated using a 10-fold, cross-validation approach, and differed only marginally for most method/trait combinations. Interestingly, for fusiform rust disease-resistance traits, Bayes Cπ, Bayes A, and RR-BLUB B had higher predictive ability than RR-BLUP and Bayesian LASSO. Fusiform rust is controlled by few genes of large effect. A limitation of RR-BLUP is the assumption of equal contribution of all markers to the observed variation. However, RR-BLUP B performed equally well as the Bayesian approaches.The genotypic and phenotypic data used in this study are publically available for comparative analysis of genomic selection prediction models.     

Predicting breeding values and genetic components using generalized linear mixed models for categorical and continuous traits in walnut (<Emphasis Type="Italic">Juglans regia</Emphasis>)

Genetic components for economically important traits in walnut (Juglans regia) were estimated for the first time using historical pedigree and heirloom phenotypic data from the walnut breeding program at the University of California, Davis. The constructed pedigree is composed of ~ 15,000 individuals and is derived from current and historic phenotypic records dating back > 50 years and located across California. To predict the additive genetic values of individuals under selection, generalized linear mixed models (GLMM), implemented with MCMCglmm, were developed. Several repeatability models were established to obtain the best model and predict the genetic parameters for each trait. Repeatability for yield, harvest date, extra-light kernel color (ELKC), and lateral bearing were predicted at 0.82, 0.98, 0.63, and 0.96, respectively, and average narrow-sense heritabilities were 0.54, 0.77, 0.49, and 0.75, respectively. Each individual in the pedigree was ranked by its estimated breeding value (EBV). The genetic trend showed specific patterns for each trait, and real genetic improvement was found over time. The completed pedigree built here, the estimated breeding values, and the ranking of individuals according to their breeding values, can be used to guide future crossing designs in the walnut breeding program and future implementation of genomic selection methods in walnut.     

Overview of LASSO-related penalized regression methods for quantitative trait mapping and genomic selection

Quantitative trait loci (QTL)/association mapping aims at finding genomic loci associated with the phenotypes, whereas genomic selection focuses on breeding value prediction based on genomic data. Variable selection is a key to both of these tasks as it allows to (1) detect clear mapping signals of QTL activity, and (2) predict the genome-enhanced breeding values accurately. In this paper, we provide an overview of a statistical method called least absolute shrinkage and selection operator (LASSO) and two of its generalizations named elastic net and adaptive LASSO in the contexts of QTL mapping and genomic breeding value prediction in plants (or animals). We also briefly summarize the Bayesian interpretation of LASSO, and the inspired hierarchical Bayesian models. We illustrate the implementation and examine the performance of methods using three public data sets: (1) North American barley data with 127 individuals and 145 markers, (2) a simulated QTLMAS XII data with 5,865 individuals and 6,000 markers for both QTL mapping and genomic selection, and (3) a wheat data with 599 individuals and 1,279 markers only for genomic selection.     

Application of Genomics Tools to Animal Breeding

The main goal in animal breeding is to select individuals that have high breeding values for traits of interest as parents to produce the next generation and to do so as quickly as possible. To date, most programs rely on statistical analysis of large data bases with phenotypes on breeding populations by linear mixed model methodology to estimate breeding values on selection candidates. However, there is a long history of research on the use of genetic markers to identify quantitative trait loci and their use in marker-assisted selection but with limited implementation in practical breeding programs. The advent of high-density SNP genotyping, combined with novel statistical methods for the use of this data to estimate breeding values, has resulted in the recent extensive application of genomic or whole-genome selection in dairy cattle and research to implement genomic selection in other livestock species is underway. The high-density SNP data also provides opportunities to detect QTL and to encover the genetic architecture of quantitative traits, in terms of the distribution of the size of genetic effects that contribute to trait differences in a population. Results show that this genetic architecture differs between traits but that for most traits, over 50% of the genetic variation resides in genomic regions with small effects that are of the order of magnitude that is expected under a highly polygenic model of inheritance.     

Deregressed EBV as the response variable yield more reliable genomic predictions than traditional EBV in pure-bred pigs

Background

Genomic selection can be implemented by a multi-step procedure, which requires a response variable and a statistical method. For pure-bred pigs, it was hypothesised that deregressed estimated breeding values (EBV) with the parent average removed as the response variable generate higher reliabilities of genomic breeding values than EBV, and that the normal, thick-tailed and mixture-distribution models yield similar reliabilities.

Methods

Reliabilities of genomic breeding values were estimated with EBV and deregressed EBV as response variables and under the three statistical methods, genomic BLUP, Bayesian Lasso and MIXTURE. The methods were examined by splitting data into a reference data set of 1375 genotyped animals that were performance tested before October 2008, and 536 genotyped validation animals that were performance tested after October 2008. The traits examined were daily gain and feed conversion ratio.

Results

Using deregressed EBV as the response variable yielded 18 to 39% higher reliabilities of the genomic breeding values than using EBV as the response variable. For daily gain, the increase in reliability due to deregression was significant and approximately 35%, whereas for feed conversion ratio it ranged between 18 and 39% and was significant only when MIXTURE was used. Genomic BLUP, Bayesian Lasso and MIXTURE had similar reliabilities.

Conclusions

Deregressed EBV is the preferred response variable, whereas the choice of statistical method is less critical for pure-bred pigs. The increase of 18 to 39% in reliability is worthwhile, since the reliabilities of the genomic breeding values directly affect the returns from genomic selection.     

Bayesian methods for estimating GEBVs of threshold traits

Estimation of genomic breeding values is the key step in genomic selection (GS). Many methods have been proposed for continuous traits, but methods for threshold traits are still scarce. Here we introduced threshold model to the framework of GS, and specifically, we extended the three Bayesian methods BayesA, BayesB and BayesCπ on the basis of threshold model for estimating genomic breeding values of threshold traits, and the extended methods are correspondingly termed BayesTA, BayesTB and BayesTCπ. Computing procedures of the three BayesT methods using Markov Chain Monte Carlo algorithm were derived. A simulation study was performed to investigate the benefit of the presented methods in accuracy with the genomic estimated breeding values (GEBVs) for threshold traits. Factors affecting the performance of the three BayesT methods were addressed. As expected, the three BayesT methods generally performed better than the corresponding normal Bayesian methods, in particular when the number of phenotypic categories was small. In the standard scenario (number of categories=2, incidence=30%, number of quantitative trait loci=50, h²=0.3), the accuracies were improved by 30.4%, 2.4%, and 5.7% points, respectively. In most scenarios, BayesTB and BayesTCπ generated similar accuracies and both performed better than BayesTA. In conclusion, our work proved that threshold model fits well for predicting GEBVs of threshold traits, and BayesTCπ is supposed to be the method of choice for GS of threshold traits.     

Response to genomic selection: The Bulmer effect and the potential of genomic selection when the number of phenotypic records is limiting

Background

Over the last ten years, genomic selection has developed enormously. Simulations and results on real data suggest that breeding values can be predicted with high accuracy using genetic markers alone. However, to reach high accuracies, large reference populations are needed. In many livestock populations or even species, such populations cannot be established when traits are difficult or expensive to record, or when the population size is small. The value of genomic selection is then questionable.

Methods

In this study, we compare traditional breeding schemes based on own performance or progeny information to genomic selection schemes, for which the number of phenotypic records is limiting. Deterministic simulations were performed using selection index theory. Our focus was on the equilibrium response obtained after a few generations of selection. Therefore, we first investigated the magnitude of the Bulmer effect with genomic selection.

Results

Results showed that the reduction in response due to the Bulmer effect is the same for genomic selection as for selection based on traditional BLUP estimated breeding values, and is independent of the accuracy of selection. The reduction in response with genomic selection is greater than with selection based directly on phenotypes without the use of pedigree information, such as mass selection. To maximize the accuracy of genomic estimated breeding values when the number of phenotypic records is limiting, the same individuals should be phenotyped and genotyped, rather than genotyping parents and phenotyping their progeny. When the generation interval cannot be reduced with genomic selection, large reference populations are required to obtain a similar response to that with selection based on BLUP estimated breeding values based on own performance or progeny information. However, when a genomic selection scheme has a moderate decrease in generation interval, relatively small reference population sizes are needed to obtain a similar response to that with selection on traditional BLUP estimated breeding values.

Conclusions

When the trait of interest cannot be recorded on the selection candidate, genomic selection schemes are very attractive even when the number of phenotypic records is limited, because traditional breeding requires progeny testing schemes with long generation intervals in those cases.     

Population structure and marker–trait associations for pomological traits in peach and nectarine cultivars

Marker–trait associations based on populations from controlled crosses have been established in peach using markers mapped on the peach consensus map. In this study, we explored the utility of unstructured populations for association mapping to determine useful marker–trait associations in peach/nectarine cultivars. We used 94 peach cultivars representing local Spanish and modern cultivars from international breeding programs that are maintained at the Experimental Station of Aula Dei, Spain. This collection was characterized for pomological traits and was screened with 40 SSR markers that span the peach genome. Population structure analysis using STRUCTURE software identified two subpopulations, the local and modern cultivars, with admixture within both groups. The local Spanish cultivars were somewhat less diverse than modern cultivars. Marker–trait associations were determined in TASSEL with and without modelling coefficient of membership (Q) values as covariates. The results showed significant associations with pomological traits. We chose three markers on LG4 because of their proximity to the endoPG locus (freestone–melting flesh) that strongly affects pomological traits. Two genotypes of BPPCT015 marker showed significant associations with harvest date, flavonoids and sorbitol. Also, two genotypes of CPPCT028 showed associations with harvest date, total phenolics, RAC, and total sugars. Finally, two genotypes of endoPG1 showed associations with flesh firmness and total sugars. The analysis of linkage disequilibrium (LD) revealed a high level of LD up to 20 cM, and decay at farther distances. Therefore, association mapping could be a powerful tool for identifying marker–trait associations and would be useful for marker-assisted selection in peach breeding.     

Genomic predictions of breeding values in a cloned <Emphasis Type="Italic">Eucalyptus globulus</Emphasis> population in Chile

In Chile, an intensive Eucalyptus globulus clonal selection program is being carried out to increase forest productivity for pulp production. A breeding population was used to investigate the predicted ability of single nucleotide polymorphism (SNP) markers for genomic selection (GS). A total of 310 clones from 53 families were used. Stem volume and wood density were measured on all clones. Trees were genotyped at 12 K polymorphic markers using the EUChip60K genotype array. Genomic best linear unbiased prediction, Bayesian lasso regression, Bayes B, and Bayes C models were used to predict genomic estimated breeding values (GEBV). For cross-validation, 260 individuals were sampled for model training and 50 individuals for model validation, using 2 folds and 10 replications each. The average predictive ability estimates for wood density and stem volume across the models were 0.58 and 0.75, respectively. The average rank correlations were 0.59 and 0.71, respectively. Models produced very similar bias for both traits. When clones were ranked based on their GEBV, models had similar phenotypic mean for the top 10% of the clones. The predicted ability of markers will likely decrease if the models are used to predict GEBV of new material coming from the breeding program, because of a different marker–trait phase introduced by recombination. The results should be validated with larger populations and across two generations before routine applications of GS in E. globulus. We suggest that GS is a viable strategy to accelerate clonal selection program of E. globulus in Chile.     

Genomic prediction applied to high-biomass sorghum for bioenergy production

The increasing cost of energy and finite oil and gas reserves have created a need to develop alternative fuels from renewable sources. Due to its abiotic stress tolerance and annual cultivation, high-biomass sorghum (Sorghum bicolor L. Moench) shows potential as a bioenergy crop. Genomic selection is a useful tool for accelerating genetic gains and could restructure plant breeding programs by enabling early selection and reducing breeding cycle duration. This work aimed at predicting breeding values via genomic selection models for 200 sorghum genotypes comprising landrace accessions and breeding lines from biomass and saccharine groups. These genotypes were divided into two sub-panels, according to breeding purpose. We evaluated the following phenotypic biomass traits: days to flowering, plant height, fresh and dry matter yield, and fiber, cellulose, hemicellulose, and lignin proportions. Genotyping by sequencing yielded more than 258,000 single-nucleotide polymorphism markers, which revealed population structure between subpanels. We then fitted and compared genomic selection models BayesA, BayesB, BayesCπ, BayesLasso, Bayes Ridge Regression and random regression best linear unbiased predictor. The resulting predictive abilities varied little between the different models, but substantially between traits. Different scenarios of prediction showed the potential of using genomic selection results between sub-panels and years, although the genotype by environment interaction negatively affected accuracies. Functional enrichment analyses performed with the marker-predicted effects suggested several interesting associations, with potential for revealing biological processes relevant to the studied quantitative traits. This work shows that genomic selection can be successfully applied in biomass sorghum breeding programs.     

Phenotypic selection for dormancy introduced a set of adaptive haplotypes from weedy into cultivated rice

Association of seed dormancy with shattering, awn, and black hull and red pericarp colors enhances survival of wild and weedy species, but challenges the use of dormancy genes in breeding varieties resistant to preharvest sprouting. A phenotypic selection and recurrent backcrossing technique was used to introduce dormancy genes from a wild-like weedy rice to a breeding line to determine their effects and linkage with the other traits. Five generations of phenotypic selection alone for low germination extremes simultaneously retained dormancy alleles at five independent QTL, including qSD12 (R(2) > 50%), as determined by genome-wide scanning for their main and/or epistatic effects in two BC(4)F(2) populations. Four dormancy loci with moderate to small effects colocated with QTL/genes for one to three of the associated traits. Multilocus response to the selection suggests that these dormancy genes are cumulative in effect, as well as networked by epistases, and that the network may have played a "sheltering" role in maintaining intact adaptive haplotypes during the evolution of weeds. Tight linkage may prevent the dormancy genes from being used in breeding programs. The major effect of qSD12 makes it an ideal target for map-based cloning and the best candidate for imparting resistance to preharvest sprouting.     

Mapping and pedigree analysis of the gene that controls the easy peel pellicle trait in Japanese chestnut (Castanea crenata Sieb. et Zucc.)

The Japanese chestnut (Castanea crenata Sieb. et Zucc.) has a pellicle that is difficult to peel, which increases the labor and cost for removing the pellicle from the nut during processing. Thus, a pellicle that is easier to peel has been an important objective of Japanese chestnut breeding programs. A newly released cultivar (“Porotan”) exhibits a unique, easily peeled pellicle. A previous study indicated that this trait is controlled by recessive gene p, and that several of the ancestors of Porotan (e.g., “Tanzawa” and 550-40) were P/p heterozygotes. Two F₁ populations from intra-specific crosses of Japanese chestnut, Tanzawa (P/p) × Porotan (p/p) and 550-40 (P/p) × Tanzawa (P/p), were used for genetic mapping of the gene that controls this characteristic. A total of 11 simple sequence repeat (SSR) markers were obtained that showed significant linkages to the p gene, and genetic linkage maps for the region around the p gene were established. Pedigree analysis was conducted for eight ancestors of Porotan around the pellicle-peeling locus using graphical genotypes based on the 11 SSR loci. The two recessive p alleles and surrounding haplotypes of Porotan were inherited through different intermediate cultivars: one allele was derived from “Otomune” (P/p) via Tanzawa and the other was derived from Otomune via Tanzawa, “Kunimi” (P/p), and breeding line 550-40. A recombination event was found in the flanking region close to the p gene in Kunimi. Molecular identification of the easy peel pellicle trait will lead to marker-assisted selection and will greatly improve Japanese chestnut breeding.     

Natural selection and inheritance of breeding time and clutch size in the collared flycatcher

Many characteristics of organisms in free-living populations appear to be under directional selection, possess additive genetic variance, and yet show no evolutionary response to selection. Avian breeding time and clutch size are often-cited examples of such characters. We report analyses of inheritance of, and selection on, these traits in a long-term study of a wild population of the collared flycatcher Ficedula albicollis. We used mixed model analysis with REML estimation ("animal models") to make full use of the information in complex multigenerational pedigrees. Heritability of laying date, but not clutch size, was lower than that estimated previously using parent-offspring regressions, although for both traits there was evidence of substantial additive genetic variance (h2 = 0.19 and 0.29, respectively). Laying date and clutch size were negatively genetically correlated (rA = -0.41 +/- 0.09), implying that selection on one of the traits would cause a correlated response in the other, but there was little evidence to suggest that evolution of either trait would be constrained by correlations with other phenotypic characters. Analysis of selection on these traits in females revealed consistent strong directional fecundity selection for earlier breeding at the level of the phenotype (beta = -0.28 +/- 0.03), but little evidence for stabilising selection on breeding time. We found no evidence that clutch size was independently under selection. Analysis of fecundity selection on breeding values for laying date, estimated from an animal model, indicated that selection acts directly on additive genetic variance underlying breeding time (beta = -0.20 +/- 0.04), but not on clutch size (beta = 0.03 +/- 0.05). In contrast, selection on laying date via adult female survival fluctuated in sign between years, and was opposite in sign for selection on phenotypes (negative) and breeding values (positive). Our data thus suggest that any evolutionary response to selection on laying date is partially constrained by underlying life-history trade-offs, and illustrate the difficulties in using purely phenotypic measures and incomplete fitness estimates to assess evolution of life-history trade-offs. We discuss some of the difficulties associated with understanding the evolution of laying date and clutch size in natural populations.     

Bayesian QTL mapping using genome-wide SSR markers and segregating population derived from a cross of two commercial F<Subscript>1</Subscript> hybrids of tomato

Key message

Using newly developed euchromatin-derived genomic SSR markers and a flexible Bayesian mapping method, 13 significant agricultural QTLs were identified in a segregating population derived from a four-way cross of tomato.

Abstract

So far, many QTL mapping studies in tomato have been performed for progeny obtained from crosses between two genetically distant parents, e.g., domesticated tomatoes and wild relatives. However, QTL information of quantitative traits related to yield (e.g., flower or fruit number, and total or average weight of fruits) in such intercross populations would be of limited use for breeding commercial tomato cultivars because individuals in the populations have specific genetic backgrounds underlying extremely different phenotypes between the parents such as large fruit in domesticated tomatoes and small fruit in wild relatives, which may not be reflective of the genetic variation in tomato breeding populations. In this study, we constructed F₂ population derived from a cross between two commercial F₁ cultivars in tomato to extract QTL information practical for tomato breeding. This cross corresponded to a four-way cross, because the four parental lines of the two F₁ cultivars were considered to be the founders. We developed 2510 new expressed sequence tag (EST)-based (euchromatin-derived) genomic SSR markers and selected 262 markers from these new SSR markers and publicly available SSR markers to construct a linkage map. QTL analysis for ten agricultural traits of tomato was performed based on the phenotypes and marker genotypes of F₂ plants using a flexible Bayesian method. As results, 13 QTL regions were detected for six traits by the Bayesian method developed in this study.

     

Efficiency of genomic selection for tomato fruit quality

Fruit quality is polygenic; each component has variable heritability and is difficult to assess. Genomic selection, which allows the prediction of phenotypes based on the whole-genome genotype, could vastly help to improve fruit quality. The goal of this study is to evaluate the accuracy of genomic selection for several metabolomic and quality traits by cross-validation and to estimate the impact of different factors on its accuracy. We analyzed data from 45 phenotypic traits and genotypic data obtained from a previous study of genetic association on a collection of 163 tomato accessions. We tested the influence of (1) the size of training population, (2) the number and density of molecular markers and (3) individual relatedness on the accuracy of prediction. The prediction accuracy of phenotypic values was largely related to the heritability of the traits. The size of training population increased the accuracy of predictions. Using 122 accessions and 5995 single nucleotide polymorphisms (SNPs) was the optimal condition. The density of markers and their numbers also affected the accuracy of the prediction. Using 2313 SNP markers distributed 0.1 cM or more apart from each other reduced the accuracy of prediction, and no gain in prediction accuracy was found when more markers were used in the model. Additionally, the more accessions were related, the more accurate were the predictions. Finally, the structure of the population negatively affected the prediction accuracy. In conclusion, the results obtained by cross-validation illustrated the effect of several parameters on the accuracy of prediction and revealed the potential of genomic selection in tomato breeding programs.     

Impact of selective genotyping in the training population on accuracy and bias of genomic selection

Estimating marker effects based on routinely generated phenotypic data of breeding programs is a cost-effective strategy to implement genomic selection. Truncation selection in breeding populations, however, could have a strong impact on the accuracy to predict genomic breeding values. The main objective of our study was to investigate the influence of phenotypic selection on the accuracy and bias of genomic selection. We used experimental data of 788 testcross progenies from an elite maize breeding program. The testcross progenies were evaluated in unreplicated field trials in ten environments and fingerprinted with 857 SNP markers. Random regression best linear unbiased prediction method was used in combination with fivefold cross-validation based on genotypic sampling. We observed a substantial loss in the accuracy to predict genomic breeding values in unidirectional selected populations. In contrast, estimating marker effects based on bidirectional selected populations led to only a marginal decrease in the prediction accuracy of genomic breeding values. We concluded that bidirectional selection is a valuable approach to efficiently implement genomic selection in applied plant breeding programs.