For tree improvement programs, these parameters help guide which traits are suitable for development as well as the selection of superior individuals for advanced generations. Selections can be made through forward or backward selection. Backward selection refers to choosing parents based on progeny performance, while forward selection refers to the selection of offspring.
Generally, early generation improvement programs practice backward selection while advanced generation programs use forward selections. Selections are based on the ranking of progeny and parents by their breeding value as a deviation from a population mean of zero White et al. One of the greatest contributions came from the development of mixed model MM theory by Henderson The key to MM analyses is that it permits the analysis of unbalanced datasets with multiple-generations and various mating designs White et al.
Further, it can account for inbreeding and deviations from panmixia White et al. The crucial difference between OLS and MM methods is the presence of both fixed and random terms in the MMs and the handling of genetic effects as random effects. This assumption allows for best linear unbiased prediction BLUP estimates of the random genetic components of the model, while fixed components are estimated using generalized least squares GLS to produce best linear unbiased estimates BLUE. BLUE and BLUP estimates have the distinct advantage over OLS estimates in that they are weighted depending on the quantity and quality of data available ensuring that the not only the best performing but the most tested individuals are ranked highest White et al.
Robust methods for estimating variance components that were suitable to the complex nature of breeding programs were also needed. The restricted maximum likelihood REML method can account for the previously described complexities of breeding programs Lynch and Walsh Likewise, modern tree improvement relies heavily on incorporating structured pedigree information in to linear mixed model analysis for prediction of individual EBVs.
The MAS strategy was primarily appealing to breeders as a method to decrease the length of the extensive tree breeding cycle, which can take upwards of 30 years for a single cycle Isik It was thought that few markers of large effect that are in LD with QTL would provide sufficient information for prediction of phenotypes. However, MAS has generally not been a rewarding endeavor in forest tree breeding programs due to several limitations of the method Strauss et al. The primary constraint that has withheld MAS from operational use in forest trees is the low proportion of the phenotypic variance accounted for by the relatively small number of statistically significant markers used in the analysis White et al.
This limitation results primarily from the infinitesimal genetic architecture i. The latter limitation constrains the transferability of estimated marker effects between environments and genetic lineages respectively, rendering phenotypic predictions to be inaccurate if violated. Genomic selection GS is fundamentally different from MAS through its simultaneous use of phenotypes and dense set of markers thousands , which are implemented without the a priori assumption concerning marker significance Meuwissen et al.
GS is thus able to capture more phenotypic variation in traits with complex inheritance since it is assumed that at least some of the very many fitted markers will be in LD with some of the QTL of the desired trait Resende et al. Thus, the novel GS approach combines the phenotypic and genetic information of a training population to develop a prediction model that produce genomic estimated breeding values GEBV for selection candidates, requiring only their genotypic information Meuwissen et al. This method has the potential to circumvent the need for the long testing phase that forest trees require to attain accurate phenotypic data for traditional pedigree-based estimation of EBVs, resulting in greater genetic gain per unit time.
This tool is especially useful in open-pollinated mating designs where the assumption of half-sib families fails often fails due to the presence of cryptic relatedness i. K-fold cross-validation is a widely used method of evaluating the prediction accuracy PA of a model Lorenz et al. The experimental population is split into K number of folds and the genomic prediction model is then trained using the phenotypic and genetic data from the K-1 folds portion of the population.
The remaining fold is used for validation of the model by calculating the average Pearson correlation coefficient between the predicted phenotype or GEBV and the corrected phenotype or an estimate of the true breeding value TBV , respectively Lorenz et al. This process is replicated to allow for the calculation of the standard error of prediction.
Ideally, to fully generalize the genomic prediction model, the restriction of factors such as environment and genetic relatedness between the training and validation population is required when assessing the PA, however, this can depend on the rationale of constructing the prediction model. Grattapaglia noted that these factors are not independent and are strongly interconnected, affecting the PA of genomic prediction models jointly.
The relationship between these two parameters essentially describes the extent of marker-QTL LD in the population. In conifers, this generally translates to a requirement of 20,, markers which is now easily obtainable through current generation sequencing and genotyping platforms Grattapaglia Additionally, training population size was presented by Grattapaglia and Resende as having little impact on genomic PA when more than individuals were used.
This relationship confirmed that decreases in PA with decreasing trait heritability can be compensated for by increasing the training population size as suggested by Meuwissen et al. An important aspect of GS to consider is the persistence of PA over subsequent generations of breeding. The theoretical increase in genetic gain produced by GS hinges on the capacity of the prediction model to remain relevant in successive generations. The PA of genomic models 7 can be composed of a combination of the two factors, leading to inflated estimates of PA to occur without proper validation since the LD between marker-QTL is reduced only by rare recombination events Habier et al.
Thus, the GS prediction models should ideally be validated after genetic recombination has occurred in a progeny generation. Two classes of statistical models are most widely used for GS in the literature, shrinkage models and variable selection models. Shrinkage models fit the effects of all markers regardless of the magnitude of their effect, whereas variable selection methods eliminate markers with effects near zero. Variable selection methods then operate under the assumption that a trait of concern is controlled by few to many loci i. Simulation studies have shown that variable selection models provide improved PA for traits with oligogenic architecture, however, results from real data do not consistently show the same trend, justifying the need to test a wide variety of statistical models Berger et al.
The models can also be further subdivided based on the assumption of marker-specific variances.
Get this edition
Homoscedastic marker variance models assume that each marker explains an equal proportion of phenotypic variance, whereas heteroscedastic effect models allow for unique marker variances. Shrinkage models with homoscedastic effect variances, such as ridge regression BLUP, fit all markers into the prediction model. Consequently, models based under 8 these assumptions have been shown to produce a greater proportion of PA based on describing genetic relationships i. Conversely, models based on heteroscedastic marker variances and variable selection have been shown to provide stronger accuracy due to marker-QTL LD Habier et al.
Nevertheless, it is recommended to use common shrinkage models as a starting point for analysis due to their robustness Lorenz et al. More recently, fifteen empirical studies have all produced promising results in regard to the acceleration of the breeding cycle for many forest tree species; namely, eucalypts Eucalyptus spp. Resende et al.
BSP Lenz et al. Chen et al. Table 1. Tree height at ages 3, 6, 10, 15, 30 and 40 years were used to construct GS prediction models and the PA between consecutive measurement years was analyzed. Variation in PA of the GS prediction models were explored with combinations of i three methods of genetic marker imputation, representing variable genetic marker densities, and ii three methods of marker effect estimation. A comparison of the relative efficiency, as the genetic gain per unit time, of GS to classical tree improvement was also performed.
This chapter represents the first use of genotyping by sequencing genetic marker discovery in a large-scale genomic tree improvement study. The ssGBLUP approach allows the incorporation of nongenotyped individuals into the GS prediction model to participate in breeding value predictions, estimation of marker effects, and estimation of genetic parameters. The aim was to assess the feasibility of reducing the genotyping effort of the required for GS to potentially increase its cost efficiency.
Genomic selection proof-of-concept in two major conifer species
The ssGBLUP approach facilitated the inclusion of phenotypic data from 7 maintained progeny trials and 7 abandoned progeny trials representing 14 environments from across the breeding zone of coastal Douglas-fir Pseudotsuga menziesii Mirb. Franco var. Additionally, monthly averages of climate variables were used to explain phenotypic variation due to environmental and genotype by environment interaction effects.
Hierarchical models with the successive inclusion of random effects to model environmental and genotype by environment variance were compared based on their genetic parameter estimates, model fit criteria, and inter- and intra-generation prediction accuracy for unobserved environments. This chapter represents the first use of environmental covariates to explain 11 environmental heterogeneity and genotype by environment interactions for the genomic prediction of breeding values in a forest tree species.
In some programs, it can take up to 30 years to complete a single cycle of breeding, specifically for traits with late expression patterns. Strategies to maximize genetic gain per unit time should then be the primary focus to rationalize the enormous spatial and economic requirements associated with forest tree-improvement practices White et al. The concept of genomic selection GS Meuwissen et al. Though this movement has yet to occur within a forest tree species context. The novel GS approach combines phenotypes and genotypes of a training population TP to develop a prediction model that estimates genomic breeding values GEBV for selection candidates, requiring only their genotypic information Meuwissen et al.
This method may circumvent the need for the long testing phase that forest trees require to attain accurate phenotypic data for traditional pedigree-based estimation of breeding values and offers a unique opportunity to substantially increase the response to selection through increasing the number of selection candidates. Previously, marker-assisted early selection MAES has been considered as a selection strategy for forest tree breeding to exploit the linkage disequilibrium LD between quantitative 16 trait loci QTL and genetic markers White et al.
However, MAES has not been rewarding in forest tree breeding programs owing to its severe limitations Strauss et al. The primary constraint that has withheld MAES from use in forest trees is the low proportion of the phenotypic variance accounted for by the relatively low number of statistically significant markers used in the analysis White et al.
This limitation results primarily from the infinitesimal genetic architecture of most complex growth-related traits Hill et al. GS is fundamentally different from MAES through its simultaneous use of phenotypes and dense set of markers thousands , which are implemented without a prior assumption concerning marker significance. GS is thus thought to capture more variation in traits with complex inheritance because it is assumed that at least some of the many fitted markers will be in LD with some of the QTL of the desired trait Meuwissen et al.
The GS method has been enabled via current generation sequencing technologies, their low per-sample cost and the use high-density single nucleotide polymorphism SNP genotyping platforms such as genotyping-by-sequencing GBS Elshire et al. The GBS method is characterized by its use of methylation-sensitive restriction enzymes to reduce genome complexity, and high levels of multiplexing, which efficiently obtain genome-wide SNP markers.
The GBS pipeline thus does not require prior genomic information, making it suitable for non-model species such as forest trees owing to their current lack of high-quality reference genomes Elshire et al. Recently, Chen et al. The density of SNP 17 markers obtained by GBS can be increased substantially by tolerating high levels of missing data and the use of marker imputation Crossa et al. However, the benefit of using imputation and the optimal imputation method has not yet been completely validated in GS studies that utilize GBS Rutkoski et al.
The potential for use of GS in forest trees was first explored by Grattapaglia and Resende through the use of deterministic simulation studies. More recently, empirical studies have all produced promising results in regard to the acceleration of the breeding cycle for three tree species; namely, eucalypts Eucalyptus spp. This study represents a novel approach over the preceding studies through the application of the non-model organism GBS SNP discovery pipeline, in addition to high missing data ratio imputation methods to produce GS prediction models.
Height at ages 3, 6, 10, 15, 30 and 40 were used to obtain estimates of pedigree-based breeding values EBV , narrow-sense heritability, age-age genetic correlations, and in combination with SNP marker data, GS prediction models and associated GEBV. The two sites are represented by the same open-pollinated families each planted in tree-row plots within 10 replicate blocks and 2. This study concerns randomly selected trees from within a subset of 25 elite families based on breeding value for tree volume. Standard errors of heritability and genetic correlation estimates were approximated using the Delta method Lynch and Walsh , using the standard error estimates for variance components obtained from ASReml.
CEF - Membres - NathalieIsabel
All analyses were completed using R software R-Core-Team The following base model was implemented Moser et al. The BLUP solution for marker effects,? In the first step, initial estimates of? However, the BLUP estimate,?
Trace plots were visually checked for model convergence. To explore the relative efficiency of GS to traditional selection TS , estimates of PA from model 1 using raw phenotypes of the genotyped individuals as training data were obtained by applying the same cross-validation method previously stated. Breeding value estimates were then obtained under two scenarios. As expected, the age-age genetic correlation between juvenile and mature height HT40 increased with increasing juvenile age Table 2.
Mean individual accuracy of breeding values estimated with the MBLUP method were consistent across years of measurement and ranged from 0. The increase in PA relative to the baseline M60 imputation method, averaged across statistical approaches and ages, was greatest for SVD 7. The former result produced a trend of diminishing return when comparing the number of markers used in GS analyses.
On average, variation in the relative difference between GS analytical approaches was less than that among imputation methods and number of markers Figure 2. Standard errors of the predictions computed from the 10 replicates were low, ranging from 0. As expected, the TPA decreased with increasing difference between the training and VP age of height measurement Table 2. Interestingly, the TPA of GS models based on year data was often equivalent to those based on year data.
Pearson product-moment correlation was used to assess the linear relationship between the absolute value of marker effects in the three analytical approaches Figure 2. Additional marker effect plots for the remaining height measurement years are available in Appendix A. PA reported in this study varied substantially throughout time Figure 2. Interestingly, the large drop in PA at age 10 and 15 years seems to coincide with a period of intense competitive exclusion between trees at this age, perhaps exacerbated by the relatively narrow tree spacing 2.
The observed extent of PA for tree height was comparable with that reported in other studies using clonal eucalypts Resende et al. More recently, Beaulieu et al. Thus, the TPA of GS methods is an important consideration for retraining said models as it offers the potential to further accelerate the breeding cycle if target phenotypes can be selected earlier. TPA in this study decreased as the difference in age of training and VP increased.
Interestingly, the TPA of GS models based on year height was nearly equivalent to those based on year despite the 10 years difference in measurements, suggesting consistency between the EBVs and SNP effects at both ages which is reflected by their high age-age genetic correlation. This analysis is in agreement with that of Resende et al.
These results are not unexpected as conifers typically have weak age-age genetic correlations attributed to their long lifespans and exposure to a wide range of environmental contingencies over time Namkoong et al. Currently, selection based on growth attributes is carried out at the age 15 years in interior spruce. The marker effect plots Figure 2.
Similarly, Resende et al. Indeed, this result has been found to be true in the present as well as other studies involving forest trees Resende et al. In theory, the statistical approach used in GS can lead to variation in PA, depending on the genetic architecture of the trait Daetwyler et al. Variable selection methods are generally expected to perform optimally for traits with simple genetic architecture that is, few loci with large effect , because SNPs of low effect are strongly shrunk toward zero, while those of large effect persist.
Beaulieu et al. As observed in the distributions of marker effects at these juvenile ages, the GRR model appeared to shrink all markers equally because of an apparent absence of those with large effect compared with mature ages Figure 2. However, in mature ages where large marker 30 effects were perceived to exist by the GRR model, the intense shrinkage applied to markers of low effect led to an obvious impairment of PA.
This is expected because of the complex genetic nature of tree height, guiding the decision that these large effect markers in mature tree height were likely false positives. This study represents the first use of GBS data as a basis for genomic prediction in a forest tree species. However, it is unclear whether this result is the product of the number of markers retained by the imputation method or the imputation method itself. Thus, the imputation method should be wholly evaluated based on both its marker yield and PA, because restricting the markers to a common set would lend unintentional penalization to methods that yield greater numbers of markers.
Similarly, Rutkoski et al. The former result alludes to diminishing returns given the average difference in available markers after filtering the TP SNP tables for a minimum minor allele frequency of 0. This result corresponds well to the asymptotic relationship between marker density and PA derived by Grattapaglia and Resende who used simulated data and deterministic formulae.
- Article Metrics;
- Next generation Sitka spruce breeding informed by predictive and comparative genomics.
- 1st Edition;
- Spatial cloud computing: a practical approach.
Foundationally, the GBS method produces a large number of missing data due to low read depth and possible mutation at restriction sites in some individuals Elshire et al. With the availability of a complete reference genome assembly of white spruce, it may be possible to increase the accuracy of imputation through use of methods designed for ordered data and constructed haplotypes Rutkoski et al.
In the interim, scaffolds of the draft genome assembly for white spruce published by Birol et al. This would be greatly beneficial because the genome size and complexity of conifers demand a large number of markers to sufficiently saturate the genome and provide high PA. The relative efficiency results are compatible with those described by other studies involving white spruce and loblolly pine Resende et al.
The theoretical increase in genetic gain produced by GS hinges on the capacity of the prediction models to remain relevant in the next generation. Recently, the source of the relationship between marker and QTL described by GS models has been decomposed by Habier et al. The origin of PA produced by the GS models in this study is currently unknown, and further testing via a progeny VP, or the partitioning of families in a restricted cross-validation scheme as demonstrated by Beaulieu et al.
Sub-optimally, PA was derived through kinship information between individuals in the training and VPs. GS PA can be composed of a combination of the two factors, leading to inflated estimates of PA to occur without proper validation because the kinship component is anticipated to decay more rapidly than marker-QTL LD in the progeny generations Habier et al.
In a study of a large population of white spruce open-pollinated families, Beaulieu et al. This is not unexpected in forest trees, where decay in LD is typically fast Neale and Savolainen Ideally, data fitted to train GS models should be analogous to the true additive genetic merit of each individual in the TP Garrick et al. However, there are empirical results in animal breeding to suggest EBV to be superior in some cases Guo et al. Kinship explained by marker data can be used to overcome this limitation and has been used in the past to correct pedigree errors Munoz et al.
Munoz et al. This concept is most relevant to young breeding programs with open-pollinated mating structures such as the one studied here. This quality is particularly appealing to tree breeders, where lengthy improvement cycles are the norm. A series of repeated tree height measurements through ages 3—40 years permitted the testing of GS methods temporally. Further, three diverse GS models were evaluated based on predictive accuracy PA , and their marker effects.
Additionally, PA varied substantially through time accordingly with spatial competition among trees. The imputation comparisons indicate that k-nearest neighbor and singular value decomposition yielded a greater number of SNPs and gave higher predictive accuracies than imputing with the mean. However, retraining of GS models will likely require phenotypic data from mid-rotation age or later to accurately reflect mature growth trait performance.
Furthermore, genotyping by sequencing was proven as a suitable marker discovery platform for the application of genomic selection in forest trees, with larger numbers of marker producing slightly greater PA. The choice of statistical model did not have a large impact on the PA, suggesting complex genetic architecture for tree height and that rrBLUP is a good baseline model for growth traits. Table 2. Standard errors in parentheses. Progeny testing is the vehicle by which these genetic parameters are estimated, while individual tree data, family composition, pedigree depth and connectedness, and field performance determine their reliability Huber et al.
Open-pollinated testing is often used to efficiently screen and assess large numbers of candidate trees Burdon and Shelbourne ; Jayawickrama and Carson The use of molecular markers can uncover hidden relatedness and potential pedigree errors in open-pollinated populations via pedigree reconstruction and paternity assignment Wang ; Kalinowski et al. The advantages of using marker-based relationship estimates are: 1 bypassing the classical mating designs needed for generating structured pedigree El-Kassaby et al. Complete details of the test site are available in Beaulieu et al.
The SNPs were located within genes and separated by a minimum distance of bp. A total of SNPs were available for use in the analyses. This process was repeated 30 times at each level of genotyping effort and mean values for the results are reported. The restricted maximum likelihood REML was used to estimate variance and covariance for the random effects in the bivariate mixed model 1 and were obtained with the ASReml-R v3. The theoretical accuracy of breeding values? A smaller AIC value indicates better fit.
The ranges of the values within the various relationship groups also tended to expand with increasing genotyping effort. These deviations imply the possible presence of pedigree errors on one side, and full-sibs within the half-sib families on the other side. Family and provenance structure was visually evident in the heat maps of the relationship estimators Appendix B, Figure B. Samples were sorted according to family and provenance, then the pairwise relationship coefficients of the matrices were plotted. Historical relationships i. Further, potential pedigree errors not accounted for in the contemporary pedigree were uncovered, these errors may be due to mislabeling or technical errors in the lab.
Further, the additive genetic component estimates of the HBLUP models continually decreased as a result of increased genotyping effort. It should be noted that the observed differences in additive genetic variance estimates between the ABLUP and HBLUP models reflect on the known bias 50 associated with considering wind-pollinated offspring as half-sib progeny, which often results in the overestimation of additive genetic variance and subsequently incorrect parental and progeny rankings.
This bias is attributable to hidden relationships and dominance effects in assumed open-pollinated half-sib pedigrees. Narrow-sense heritability estimates of the bivariate HBLUP models mirrored those of the additive genetic variance estimates HT: from 0. The trend observed in the estimates of additive genetic variance was coupled with an increase in the residual variance component for the HBLUP models compared to the ABLUP model, with the lowest estimates observed for the ABLUP model, indicating that some residual variance was shifted to the additive variance, hence the overestimation of heritability Table 3.
Comparison of the block variance estimates provided no pattern of association with any factor, and the estimates were relatively stable across all models for both WD and HT. Across all the models, the additive genetic correlations between HT and WD were minor, negative, and not significant based on SEs.
The pairwise Spearman rank correlations were always less than perfect indicating at least some disagreement in the rankings of candidate maternal parents and progeny. Commonly, quantitative genetics analyses are based on the animal model implemented in mixed model theory Henderson The mixed model theory operates with the assumption that variance—covariance matrices of random terms used in the animal model are error free Henderson The average numerator relationship matrix cannot fulfill this assumption as it is based on the pairwise expected relationship values and does not account for the Mendelian sampling term.
Moreover, the contemporary shallow pedigrees of many forest tree breeding programs are unable to account for historical coancestry. The use of genotypic data enables greater insight into genetic covariances and the precise mapping of the Mendelian sampling term Visscher et al.
Additionally, deployment of genetic analysis using the entire testing population i. While the genomic information benefits from lower inbreeding build-up during the selection phases as compared to pedigree-based BLUP Liu et al. This is increasingly important, especially in forest tree breeding programs, which are based on shallow pedigree-based selection from large base populations. By permitting both pedigree and marker information in a single genetic evaluation analysis, the combined HBLUP approach offers resolution to these problems.
Thus, more accurate estimates of relatedness are obtained in a cost-efficient manner. Our results support this notion through large improvement in the model fit as represented by the AIC statistics from the exclusive pedigree-based model to models utilizing the combination of marker and pedigree information at various genotyping efforts Table 3. Furthermore, the presence of 54 hidden relatedness, particularly in open-pollinated trials, can be a serious problem in their genetic evaluation resulting in upwardly biased estimates of heritability Squillace ; Namkoong et al.
This bias is caused by overestimation of additive genetic variance and hidden dominance effects due to the unrealistic assumption of pure half-sibling relatedness within the open-pollinated families, as well as complete unrelatedness among the parental donors. Reality and empirical evidence suggests the existence of hidden relationships within these families i. Our results confirm that hidden relatedness within the OP families and historical coancestry exists in this P. It should be stated that all hidden relatedness can only be accounted for under the GBLUP analysis, where individuals are genotyped including parents ; however, the HBLUP is composed of a mixture of genotyped and nongenotyped individuals, thus a slight overestimation of the additive genetic variance is expected.
The accuracy of the estimated breeding values is of practical importance to tree breeders. Correct ranking of the selection candidates based on accurate estimates of relatedness is an important factor. While traditional pedigree-based analyses have been proven to deliver increased genetic gain, a reduction in the size of the testing population can be achieved with an improvement in accuracy through the use of genetic markers in the evaluation.
The comparison of breeding value accuracies in this study was based on the assumption that the variance component estimates 55 produced by the bivariate HBLUP model with full genotyping effort were most accurate. The variance components of each model were fixed to these estimates prior to calculating theoretical breeding value accuracy for each scenario. Our results show incremental improvement in the accuracy of progeny breeding values for the genotyped subset for both HT and WD with increasing genotyping effort Table 3.
However, this improvement was not translated to the nongenotyped progeny subset in either trait. The simple, disconnected pedigree structure of this population is a probable explanation for this observation, as the genomic information is only transferred from genotyped to nongenotyped individuals via pedigree relationships.
It is expected that in systems with more complex and interconnected pedigree structure, such as diallel mating designs, multi-generational complex pedigrees will benefit more from the HBLUP method as genomic information can be inferred from multiple family sources to the nongenotyped individuals. The difference in contrasting heritabilities of the traits and these findings agree with previous discussions that GS is expected to be more efficient in low heritability traits Calus et al.
Yet, the inclusion of even a small proportion i. Generally, the genomic relationship matrix uses actual allele frequencies instead of those of the base population, which are 56 usually unknown in this case, the base population is treated as the studied pedigree population , and sets the genomic-base population as the genotyped population Oliehoek et al. However, such analysis is affected by incompatibility between pedigree- and marker-based relationship matrices due to the inferences made about the base population using information from the studied population.
Powell et al. Such a matrix is efficient when the population is inbreeding free, and if inbreeding is present then inbreeding coefficients have to be included in the denominator Forni et al. Additionally, Vitezica et al. As a preliminary investigation into the combined use of genomic and pedigree information in the applied genetic analysis of white spruce, the HBLUP method has proven to be a beneficial 57 tool to forest tree breeders.
The inclusion of nongenotyped and genotyped trees in a single analysis produced improvements in breeding value accuracy and model fit, particularly for the trait with low heritability HT. The effect of the combined use of contemporary pedigree information and genomic relatedness estimates on the accuracy of predicted breeding values and precision of estimated genetic parameters, as well as rankings of selection candidates, using single-step genomic evaluation HBLUP was investigated. The addition of genomic information in the analysis considerably improved the accuracy in breeding value estimates by accounting for both Mendelian sampling and historical coancestry that were not captured by the contemporary 58 pedigree alone.
Increasing within family genotyping efforts were associated with continuous improvement in model fit, precision of genetic parameters, and breeding value accuracy. Yet, improvements were observed even at minimal genotyping effort, indicating that even modest genotyping effort is effective in improving genetic evaluation. Even modest genotyping effort within open-pollinated families is effective at improving important population level genetic parameters such as heritability and genetic correlations.
Further, the modest within family genotyping effort was effective at capturing historical coancestry i. In the broader sense, this chapter illustrates that HBLUP the combined utilization of both pedigree and genomic information is a cost-effective approach to increase the accuracy of breeding values in forest tree breeding programs where shallow pedigrees and large testing populations are the norm.
GE Estimator? See text for variance parameter definitions. Genomic selection GS can address some of these shortcomings through early prediction of phenotypes based on large numbers of jointly considered genomic markers, most commonly, single nucleotide polymorphisms SNPs Meuwissen et al. This solution is realized through improved management of co-ancestry and increased genetic gain via enhanced precision and accuracy of pairwise kinship estimates for the breeding population.
The recent exploration of GS within the forest tree breeding framework has produced numerous promising studies, indicating that GS prediction accuracies are able to at least match and often surpass pedigree-based predictions see reviews by Grattapaglia ; Grattapaglia et al. Therefore, it is commonplace to regionalize tree breeding efforts based on available biogeoclimatic information to avoid maladaptation of deployed stock i. However, for large numbers of environments, typically, this approach is computationally not practical and simplifications of covariance structures or factor analytic models must be put in place Isik et al.
Consequently, estimated marker effects would be considered specific to the environment s of the training populations. Without phenotypic observations from the target environment, genomic prediction is challenging. Single-step genomic evaluation ssGBLUP is a unified approach that allows the incorporation of phenotypic, genomic, and pedigree information into a single analysis Legarra et 66 al. This methodology allows the prediction of breeding values for genotyped and non-genotyped individuals to be on the same scale, avoiding bias and complex multi-step analyses Vitezica et al.
It also allows for the phenotypes of non-genotyped individuals to participate in the estimation of marker effects, effectively boosting the accuracy of prediction. Thus, the method also provides a cost-effective entry into GS as relatively few important individuals can be genotyped while phenotypic records of rogued trials can also easily be implemented yielding a modern analysis for estimating marker effects or breeding values.
Wang et al. Lourenco et al. The indirect prediction approach improves computational efficiency and allows for fast prediction of GEBV for new genotyped trees via SNP marker effects as opposed to a full ssGBLUP evaluation where new genotyped individuals need to be explicitly included. The CDF breeding program is the most advanced in BC and is currently in its third generation with advanced generation seed orchards producing 6. Using 14 experimental trials planted in different environments, monthly averages of ECs obtained from ClimateBC Wang et al.
The method of Lourenco et al. The parental P0 generation consists of 78 wild plus-tree selections, which were crossed in partial disconnected diallels to produce full-sib families for the second generation F1. A total of 31, tree height [cm] phenotypic measurements were available for ages 12 F1 and 11 F and F years. The relatedness restriction resulted in an average additive genetic kinship coefficient of 0. The SNPs used here differed from those used by Thistlethwaite et al.
Filtering criteria 69 was done using VCFtools Danecek et al. Monthly ECs were averaged across the growing period for each trial, from planting year until the year of phenotypic measurement, and included primary measures of temperature, precipitation, and solar radiation see Table C.
Dixson Library. Hawkesbury Campus Library. Open to the public Book; Illustrated English Show 0 more libraries None of your libraries hold this item. Found at these bookshops Searching - please wait We were unable to find this edition in any bookshop we are able to search. These online bookshops told us they have this item:. Tags What are tags? Add a tag. Public Private login e. Add a tag Cancel Be the first to add a tag for this edition. Lists What are lists? Login to add to list. Be the first to add this to a list.
Pages Received 05 Jan Additional information Acknowledgements An early version of this manuscript was read and commented by Matti Haapanen and Ola Rosvall, whom I am thanking for their valuable comments. Article Metrics Views. Article metrics information Disclaimer for citing articles. Login options Log in. Username Password Forgot password?
Shibboleth OpenAthens. Restore content access Restore content access for purchases made as guest. Article Purchase - Online Checkout. Issue Purchase - Online Checkout. People also read article commentary.
Related Genetics, genomics, and breeding of conifers
Copyright 2019 - All Right Reserved