Genomic Selection for Yield and Seed Protein Content in Soybean: A Study of Breeding Program Data and Assessment of Prediction Accuracy

Duhnen, Alexandra; Gras, Amandine; Teyssèdre, Simon; Romestant, Michel; Claustres, Bruno; Daydé, Jean; Mangin, Brigitte

doi:10.2135/cropsci2016.06.0496

Cited by 70 publications

(65 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Duhnen et al (2017) found similar results using G-BLUP cross-validations to the values reported in this study for yield (0.49) and protein (0.67) (Table 3), with prediction accuracies for yield ranging from 0.45 to 0.63 and for protein from 0.45 to 0.59. In this study, four selection methods (PS, BayesB, G-BLUP, and Epistacy) were evaluated for soybean yield, oleic acid, linolenic acid, protein, and oil.…”

Section: Discussionsupporting

confidence: 86%

“…This is commonly done by subsetting a portion of the population to serve as a training set, predicting the performance of another portion of the population, and then evaluating the predictions using cross-validations (Duhnen et al, 2017). Many GS studies have sought to evaluate predictions in one growing season, without testing the performance of predictions over time.…”

Section: Selection Methods and Statistical Analysismentioning

confidence: 99%

“…Many GS studies have sought to evaluate predictions in one growing season, without testing the performance of predictions over time. This is commonly done by subsetting a portion of the population to serve as a training set, predicting the performance of another portion of the population, and then evaluating the predictions using cross-validations (Duhnen et al, 2017). In contrast, we sought to evaluate the accuracy of predictions over multiple generations by making predictions using the whole 860-RIL population from 2010 and validating in a subset of the population using 276 RILs grown in replicated field trials in 2013.…”

Section: Selection Methods and Statistical Analysismentioning

confidence: 99%

See 2 more Smart Citations

Context‐Specific Genomic Selection Strategies Outperform Phenotypic Selection for Soybean Quantitative Traits in the Progeny Row Stage

et al. 2019

View full text Add to dashboard Cite

Evaluating different breeding selection strategies for relative utility is necessary to choose those that maximize efficiency. Soybean [Glycine max (L.) Merr.] seed yield and fatty acid, protein, and oil contents are all commercially important traits that display complex quantitative inheritance. A soybean population consisting of 860 F5–derived recombinant inbred lines (RILs), genotyped with 4867 polymorphic single nucleotide polymorphism (SNPs) was used to compare phenotypic and context specific genomic selection (GS) strategies. To simulate progeny rows, each RIL was grown in a single plot in 2010 in Knoxville, TN, and phenotype was recorded. A subset of 276 RILs with similar maturity was then grown in multilocation, replicated field trials in 2013 to compare the performance of each selection method in field conditions. Notably, the preferred method for each trait was GS. Of the GS approaches evaluated, Epistacy performed best for yield, and BayesB and/or genomic best linear unbiased prediction (G‐BLUP) were preferred for each of the other traits. Yield was the only trait for which the predictions had a large change when the number of SNPs and the number of RILs were randomly reduced for the G‐BLUP model, with the best predictions occurring when RILs with different maturity that were not grown in 2013 were removed from the training set. These findings provide important information on how soybean breeders can maximize selections from the progeny row stage for yield and fatty acid, protein, and oil contents by using appropriate selection strategies.

show abstract

Section: Discussionsupporting

confidence: 86%

Section: Selection Methods and Statistical Analysismentioning

confidence: 99%

Section: Selection Methods and Statistical Analysismentioning

confidence: 99%

See 1 more Smart Citation

Context‐Specific Genomic Selection Strategies Outperform Phenotypic Selection for Soybean Quantitative Traits in the Progeny Row Stage

et al. 2019

View full text Add to dashboard Cite

show abstract

“…With both methods (RR and TP), we observed high prediction accuracies ranging from 0.4 to 0.6 (Arruda et al., ; Duhnen et al., ; Kristensen et al., ; Leplat, Jensen, & Madsen, ). While similar accuracies have been reported by other studies for complex traits, it is important to note that the current study utilized a diversity panel with considerable population stratification and the prediction models did not account for population structure.…”

Section: Discussionmentioning

confidence: 92%

Utilizing random regression models for genomic prediction of a longitudinal trait derived from high‐throughput phenotyping

2018

View full text Add to dashboard Cite

The accessibility of high‐throughput phenotyping platforms in both the greenhouse and field, as well as the relatively low cost of unmanned aerial vehicles, has provided researchers with an effective means to characterize large populations throughout the growing season. These longitudinal phenotypes can provide important insight into plant development and responses to the environment. Despite the growing use of these new phenotyping approaches in plant breeding, the use of genomic prediction models for longitudinal phenotypes is limited in major crop species. The objective of this study was to demonstrate the utility of random regression ( RR ) models using Legendre polynomials for genomic prediction of shoot growth trajectories in rice ( Oryza sativa ). An estimate of shoot biomass, projected shoot area ( PSA ), was recorded over a period of 20 days for a panel of 357 diverse rice accessions using an image‐based greenhouse phenotyping platform. A RR that included a fixed second‐order Legendre polynomial, a random second‐order Legendre polynomial for the additive genetic effect, a first‐order Legendre polynomial for the environmental effect, and heterogeneous residual variances was used to model PSA trajectories. The utility of the RR model over a single time point ( TP ) approach, where PSA is fit at each time point independently, is shown through four prediction scenarios. In the first scenario, the RR and TP approaches were used to predict PSA for a set of lines lacking phenotypic data. The RR approach showed a 11.6% increase in prediction accuracy over the TP approach. Much of this improvement could be attributed to the greater additive genetic variance captured by the RR approach. The remaining scenarios focused forecasting future phenotypes using a subset of early time points for known lines with phenotypic data, as well new lines lacking phenotypic data. In all cases, PSA could be predicted with high accuracy ( r : 0.79 to 0.89 and 0.55 to 0.58 for known and unknown lines, respectively). This study provides the first application of RR models for genomic prediction of a longitudinal trait in rice and demonstrates that RR models can be effectively used to improve the accuracy of genomic prediction for complex traits compared to a TP approach.

show abstract

“…With both methods (RR and TP), we observed high prediction accuracies ranging from 0.4 to 0.6 (Arruda et al, 2015;Duhnen et al, 2017;Kristensen et al, 2018;Leplat, Jensen, & Madsen, 2016).…”

Section: Advantages Of Rr Over Univariate Genomic Predictionmentioning

confidence: 95%

Utilizing random regression models for genomic prediction of a longitudinal trait derived from high-throughput phenotyping

Campbell

Walia

Morota

2018

Preprint

View full text Add to dashboard Cite

The accessibility of high-throughput phenotyping platforms in both the greenhouse and field, as well as the relatively low cost of unmanned aerial vehicles, has provided researchers with an effective means to characterize large populations throughout the growing season. These longitudinal phenotypes can provide important insight into plant development and responses to the environment. Despite the growing use of these new phenotyping approaches in plant breeding, the use of genomic prediction models for longitudinal phenotypes is limited in major crop species. The objective of this study was to demonstrate the utility of random regression (RR) models using Legendre polynomials for genomic prediction of shoot growth trajectories in rice (Oryza sativa). An estimate of shoot biomass, projected shoot area (PSA), was recorded over a period of 20 days for a panel of 357 diverse rice accessions using an image-based greenhouse phenotyping platform. A RR that included a fixed second-order Legendre polynomial, a random second-order Legendre polynomial for the additive genetic effect, a first-order Legendre polynomial for the environmental effect, and heterogeneous residual variances was used to model PSA trajectories. The utility of the RR model over a single time point (TP)approach, where PSA is fit at each time point independently, is shown through four prediction scenarios. In the first scenario, the RR and TP approaches were used to predict PSA for a set of lines lacking phenotypic data. The RR approach showed a 11.6% increase in prediction accuracy over the TP approach. Much of this improvement could be attributed to the greater additive genetic variance captured by the RR approach. The remaining scenarios focused forecasting future phenotypes using a subset of early time points for known lines with phenotypic data, as well new lines lacking phenotypic data.In all cases, PSA could be predicted with high accuracy (r: 0.79 to 0.89 and 0.55 to 0.58 for known and unknown lines, respectively). This study provides the first application of RR models for genomic prediction of a longitudinal trait in rice and demonstrates that RR models can be effectively used to improve the accuracy of genomic prediction for complex traits compared to a TP approach. K E Y W O R D Sgenetics, genomic prediction, high-throughput phenotyping, phenomics This manuscript has been released as a preprint on bioRxiv (https://doi.

show abstract

Genomic Selection for Yield and Seed Protein Content in Soybean: A Study of Breeding Program Data and Assessment of Prediction Accuracy

Cited by 70 publications

References 36 publications

Context‐Specific Genomic Selection Strategies Outperform Phenotypic Selection for Soybean Quantitative Traits in the Progeny Row Stage

Context‐Specific Genomic Selection Strategies Outperform Phenotypic Selection for Soybean Quantitative Traits in the Progeny Row Stage

Utilizing random regression models for genomic prediction of a longitudinal trait derived from high‐throughput phenotyping

Utilizing random regression models for genomic prediction of a longitudinal trait derived from high-throughput phenotyping

Contact Info

Product

Resources

About