Search claims · PlantClaims

Method / resourceobservedoriginal resultpositivemeasured

Claim · model-written summary

Deep learning models outperformed Bayesian models across single and multiple trait and environment schemes.

Taxon: maizeZea mayst0
Outcome: outperformed Bayesian modelse0
Direction: outperformede0
Comparator: Bayesian modelse0

Verbatim

e0
“deep learning models consistently outperformed Bayesian models in both single and multiple trait and environment approaches”
✓ located verbatim in the source · abstract, characters 1271–1394

t0
“flowering traits in maize”
✓ located verbatim in the source · title, characters 57–82

Title

Multi-trait and multi-environment genomic prediction for flowering traits in maize: a deep learning approach

Freddy Mora-Poblete; Carlos Maldonado; Luma Henrique; et al. Frontiers in Plant Science 2023. doi:10.3389/fpls.2023.1153040

Other claims from this paper (7)

Method / resource Multi-trait models increased prediction accuracy 14.4% over a single-trait single-environment approach.
Method / resource A single trait in multiple environments improved accuracy 6.4% compared with multi-trait analysis.
Observation GWAS identified 26 candidate genes related to flowering time.
Observation GWAS identified 31 marker-trait associations.
Observation The identified associations accounted for 37% of ASI variation.
Observation The identified associations accounted for 37% of FF variation.
Observation The identified associations accounted for 22% of MF variation.

Method / resourceobservedoriginal resultpositivemeasured

Claim · model-written summary

Under strong epistasis, convolutional neural networks outperformed linear Bayesian models in predictive accuracy.

Taxon: strawberry and blueberrye0
Outcome: predictive accuracye3
Qualifier: strong epistasise2
Comparator: Linear Bayesian modelse2

Verbatim

e0
“two important small fruits or berries: strawberry and blueberry”
✓ located verbatim in the source · abstract, characters 800–863

e2
“Linear Bayesian models were better than convolutional neural networks for the full additive architecture, whereas the opposite was observed under strong epistasis”
✓ located verbatim in the source · abstract, characters 1630–1792

e3
“Bayesian linear models can match or exceed the predictive accuracy of DL”
✓ located verbatim in the source · abstract, characters 1888–1960

Title

Exploring Deep Learning for Complex Trait Genomic Prediction in Polyploid Outcrossing Species

Laura M. Zingaretti; Salvador Alejandro Gezan; Luis Felipe V. Ferrão; et al. Frontiers in Plant Science 2020. doi:10.3389/fpls.2020.00025

Other claims from this paper (4)

Method / resourceobservedoriginal resultpositivemeasured

Claim · model-written summary

Deep learning outperformed RRBLUP and Bayesian models in all prediction scenarios.

Taxon: Winter WheatTriticum aestivumt0
Outcome: prediction accuracies for different traitse1
Direction: superiore0
Comparator: traditional ridge regression best linear unbiased prediction (RRBLUP) and Bayesian modelse0

Verbatim

e0
“Deep learning models were superior over the traditional ridge regression best linear unbiased prediction (RRBLUP) and Bayesian models under all prediction scenarios”
✓ located verbatim in the source · abstract, characters 1407–1571

e1
“prediction accuracies for different traits”
✓ located verbatim in the source · abstract, characters 1083–1125

t0
“Winter Wheat”
✓ located verbatim in the source · title, characters 74–86

Title

Genomic Selection for End-Use Quality and Processing Traits in Soft White Winter Wheat Breeding Program with Machine and Deep Learning Models

Karansher Sandhu; Meriem Aoun; Craig Morris; et al. Biology 2021. doi:10.3390/biology10070689

Other claims from this paper (6)

Method / resource Cross-validation prediction accuracy across quality traits ranged from 0.45 to 0.81.
Method / resource Forward prediction accuracy ranged from 0.29 to 0.55.
Method / resource Across-location prediction accuracy ranged from 0.27 to 0.50.
Method / resource Forward accuracy increased over time with more training data, especially for machine and deep learning.
Recommendation The authors supported prediction of quality traits in early generations to advance superior genotypes.
Recommendation The authors supported using machine and deep learning in large breeding programs to predict complex traits.

Method / resourceobservedoriginal resultpositivemeasured

Claim · model-written summary

The deep learning model outperformed traditional models for genomic breeding value prediction.

Outcome: model prediction accuracye0
Direction: outperformse0
Comparator: traditional statistical RR-BLUP, Bayesian LASSO and BayesA modelse0

Verbatim

e0
“Our model prediction accuracy, which was calculated by Pearson Correlation between GEBV and observed values, outperforms traditional statistical RR-BLUP, Bayesian LASSO and BayesA models”
✓ located verbatim in the source · abstract, characters 1125–1311

Title

Application of deep learning in genomic selection

Yang Liu; Duolin Wang 2017 IEEE International Conference on Bioinformatics and Biomedicine (BIBM) 2017. doi:10.1109/bibm.2017.8218025

Method / resourceobservedoriginal resultpositivemeasured

Claim · model-written summary

BayesRR outperformed seven other genomic prediction models.

Taxon: wheatTriticum aestivumt0
Outcome: outperformede0
Direction: outperformede0
Comparator: other seven modelse0

Verbatim

e0
“The Bayes ridge regression (BayesRR) model outperformed the other seven models, whereas BayesLASSO (BayesL) was the least efficient”
✓ located verbatim in the source · abstract, characters 748–879

t0
“wheat”
✓ located verbatim in the source · title, characters 95–100

Title

Evaluation of eight Bayesian genomic prediction models for three micronutrient traits in bread wheat ( Triticum aestivum L.)

Prabina Kumar Meher; Ajit Gupta; Sachin Rustgi; et al. The Plant Genome 2023. doi:10.1002/tpg2.20332

Other claims from this paper (7)

Method / resource BayesLASSO was the least efficient model.
Association Genomic prediction accuracy increased with training set size.
Association Genomic prediction accuracy increased with marker density.
Method / resource Genomic prediction accuracy differed among the three micronutrient traits.
Method / resource Prediction accuracy from BLUE was higher than accuracy for the two environments.
Method / resource Prediction accuracy remained unaffected after accounting for population structure.
Recommendation The authors recommend using only the best model to estimate GEBVs before genomic selection for grain micronutrient improvement.

Method / resourceobservedoriginal resultpositivemeasured

Claim · model-written summary

Both models outperformed the provincial N recommendation system.

Taxon: cornZea mays subsp. mayst0
Outcome: outperformede0
Direction: outperformede0
Comparator: benchmark provincial N recommendation systeme0

Verbatim

e0
“Both models incorporated weather from planting to sidedress and outperformed a benchmark provincial N recommendation system.”
✓ located verbatim in the source · abstract, characters 1098–1222

t0
“EONR for corn.”
✓ located verbatim in the source · abstract, characters 414–428

Title

Comparing Random Forest to Bayesian Networks as nitrogen management decision support systems

John Sulik; Kamaljit Banger; Ken Janovicek; et al. Agronomy Journal 2023. doi:10.1002/agj2.21320

Other claims from this paper (3)

Method / resource The Bayesian Network achieved a 64% success rate in predicting corn EONR.
Method / resource The Random Forest achieved a 48% success rate in predicting corn EONR.
Method / resource The authors considered Bayesian Networks more flexible for stakeholder input and uncertainty.

Method / resourceobservedoriginal resultpositivemeasured

Claim · model-written summary

GBLUP slightly outperformed other genomic selection models.

Taxon: PeaLathyrus oleraceust0
Outcome: performed slightly bettere0
Direction: slightly bettere0
Comparator: other models trialede0

Not reportedValue

Verbatim

e0
“GBLUP and Bayesian Reproducing kernel Hilbert spaces regression (RKHS) performed slightly better than the other models trialed”
✓ located verbatim in the source · abstract, characters 1117–1243

t0
“Pea”
✓ located verbatim in the source · title, characters 53–56

Title

Genomic Selection for Ascochyta Blight Resistance in Pea

Margaret A. Carpenter; David S. Goulden; Carmel J. Woods; et al. Frontiers in Plant Science 2018. doi:10.3389/fpls.2018.01878

Other claims from this paper (6)

Method / resource The best pea ascochyta score prediction accuracy was 0.56 using GBLUP, a mean score and a 70% data quality threshold.
Method / resource RKHS slightly outperformed other genomic selection models.
Method / resource Missing-data thresholds changed prediction accuracy minimally.
Method / resource Prediction accuracy varied greatly among random training and testing partitions.
Method / resource Marker by environment interactions did not improve accuracy for unphenotyped lines.
Method / resource Marker by environment interactions improved prediction across environments.

Method / resourceobservedoriginal resultpositivemeasured

Claim · model-written summary

RKHS slightly outperformed other genomic selection models.

Taxon: PeaLathyrus oleraceust0
Outcome: performed slightly bettere0
Direction: slightly bettere0
Comparator: other models trialede0

Verbatim

e0
“GBLUP and Bayesian Reproducing kernel Hilbert spaces regression (RKHS) performed slightly better than the other models trialed”
✓ located verbatim in the source · abstract, characters 1117–1243

t0
“Pea”
✓ located verbatim in the source · title, characters 53–56

Title

Genomic Selection for Ascochyta Blight Resistance in Pea

Margaret A. Carpenter; David S. Goulden; Carmel J. Woods; et al. Frontiers in Plant Science 2018. doi:10.3389/fpls.2018.01878

Other claims from this paper (6)

Method / resource The best pea ascochyta score prediction accuracy was 0.56 using GBLUP, a mean score and a 70% data quality threshold.
Method / resource GBLUP slightly outperformed other genomic selection models.
Method / resource Missing-data thresholds changed prediction accuracy minimally.
Method / resource Prediction accuracy varied greatly among random training and testing partitions.
Method / resource Marker by environment interactions did not improve accuracy for unphenotyped lines.
Method / resource Marker by environment interactions improved prediction across environments.

Observationobservedoriginal resultpositivemeasured

Claim · model-written summary

G4 consistently outperformed other candidates across stability parameters.

Taxon: barleyHordeum vulgaret0
Material: G4e0
Outcome: outperformed other candidatese0
Direction: outperformede0

Verbatim

e0
“genotype G4, derived from the pedigree [Sahra/3/Bda/Rhn-03//ICB-107766], consistently outperformed other candidates”
✓ located verbatim in the source · abstract, characters 1696–1811

t0
“Integrating AMMI, Bayesian-AMMI, BLUP, and GGE biplot models to improve the identification of high-yielding and stable barley genotypes”
✓ located verbatim in the source · title, characters 0–135

Title

Integrating AMMI, Bayesian-AMMI, BLUP, and GGE biplot models to improve the identification of high-yielding and stable barley genotypes

Alireza Pour-Aboughadareh; Hassan Zali; Ahmad Gholipour; et al. BMC Plant Biology 2026. doi:10.1186/s12870-026-09142-0

Other claims from this paper (9)

Method / resourceobservedoriginal resultpositivemeasured

Claim · model-written summary

Species-effect HTT models outperformed models without a species effect.

Outcome: superiore0
Direction: superiore0
Comparator: those that did note0

Not reportedTaxon

Verbatim

e0
“HTT models that accounted for a species effect were superior to those that did not.”
✓ located verbatim in the source · abstract, characters 1002–1085

Title

Comparison of three different statistical approaches (non-linear least-squares regression, survival analysis and Bayesian inference) in their usefulness for estimating hydrothermal time models of seed germination

Elena Moltchanova; Shirin Sharifiamina; Derrick J. Moot; et al. Seed Science Research 2020. doi:10.1017/s0960258520000082

Other claims from this paper (5)

Method / resource All three methods estimated similar parameters and root mean squared errors.
Method / resource Bayesian estimation was efficient without computation-intensive methods.
Method / resource Bayesian estimation allowed easy comparison of species parameters.
Association Most HTT parameters credibly differed between the two clover species.
Association Species parameter differences agreed with known differences in germination behaviour.

“outperformed Bayesian models”

Claim types

Direction of the findings

Over time

Claim type

Research area

Polarity

Statistics