Search claims · PlantClaims

Method / resourceobservedoriginal resultpositivemeasured

Claim · model-written summary

The ensemble yew model had high predictive performance.

Taxon: Yewt0
Outcome: predictive performancee0
Value: high predictive performancee0

Verbatim

t0
“Yew”
✓ located verbatim in the source · title, characters 58–61

e0
“The ensemble model showed high predictive performance”
✓ located verbatim in the source · abstract, characters 1174–1227

Title

Climate Vulnerability Analysis of Marginal Populations of Yew (Taxus baccata L.): The Case of the Iberian Peninsula

Jhony Fernando Cruz Román; Ricardo Enrique Hernández-Lambraño; David Rodríguez-de la Cruz; et al. Forests 2025. doi:10.3390/f16060931

Other claims from this paper (4)

Method / resourceobservedoriginal resultmeasured

Claim · model-written summary

Xception had moderate predictive performance.

Taxon: Plantt0
Outcome: predictive performancee0

Not reportedComparatorValue

Verbatim

e0
“Xception performs moderately in predictive performance”
✓ located verbatim in the source · abstract, characters 1482–1536

t0
“Plant Disease Detection Models”
✓ located verbatim in the source · title, characters 24–54

Title

Comparative Analysis Of Plant Disease Detection Models: Toward Sustainable And Environmental-Friendly Agriculture

Sonali Patil; Manisha Mali; Pushpi Rani; et al. International Journal of Environmental Sciences 2025. doi:10.64252/kwz45z64

Other claims from this paper (3)

Method / resource The ExtraTrees and KNN hybrid had the highest overall efficiency with strong accuracy and minimal overhead.
Method / resource Vision Transformer was competitive in accuracy but used more resources.
Method / resource ResNet-50 and EfficientNet-B5 performed well.

Method / resourceobservedoriginal resultmeasured

Claim · model-written summary

SVR had superior predictive performance.

Material: SVRe0
Outcome: superior predictive performancee0

Not reportedTaxonComparator

Verbatim

e0
“Among these, SVR, LGB, and DNNGP demonstrate superior predictive performance.”
✓ located verbatim in the source · abstract, characters 4708–4785

t0
“peach”
✓ located verbatim in the source · title, characters 85–90

Title

PeachDB: accelerating breeding by reducing cost with AI-powered genomic selection in peach

Qingyuan Han; Yuhao Lou; Tianyu Zhu; et al. Horticulture Research 2026. doi:10.1093/hr/uhag040

Other claims from this paper (19)

Method / resource PeachDB provides 246 genomic selection models.
Method / resource The citric acid prediction accuracy was 0.74.
Method / resource The malic acid prediction accuracy was 0.71.
Method / resource The quinic acid prediction accuracy was 0.57.
Method / resource LGB had superior predictive performance.
Method / resource DNNGP had superior predictive performance.
Method / resource The citric acid model reached only 0.22 maximum accuracy in the hybrid population.
Method / resource All evaluated malic acid models exceeded 0.7 accuracy in the hybrid population.
Method / resource The highest malic acid model accuracy reached 0.83.
Association Five molecular markers correlated with organic acid content variation.
Observation Fifty-nine of 155 hybrid offspring carried the low-acid genotype at all five markers.
Intervention effect In 2018 GS selected fruit had 16.5% lower mean total acid content than MAS selected fruit.
Intervention effect In 2019 GS selected fruit had 13.13% lower mean total acid content than MAS selected fruit.
Intervention effect Overall GS selected fruit had 14.12% lower mean total acid content than MAS selected fruit.
Method / resource GS screening required 86% fewer low-acid individuals for retention than MAS.
Method / resource GS reduced genotyping cost per hybrid offspring by 50% compared with MAS.
Method / resource GS reduced evaluation land area by 86%.
Method / resource GS improved phenotypic performance by 15%.
Method / resource PeachDB makes genomic resources and downloadable selection models publicly accessible.

Method / resourceobservedoriginal resultmeasured

Claim · model-written summary

LGB had superior predictive performance.

Material: LGBe0
Outcome: superior predictive performancee0

Not reportedTaxonComparator

Verbatim

e0
“Among these, SVR, LGB, and DNNGP demonstrate superior predictive performance.”
✓ located verbatim in the source · abstract, characters 4708–4785

t0
“peach”
✓ located verbatim in the source · title, characters 85–90

Title

PeachDB: accelerating breeding by reducing cost with AI-powered genomic selection in peach

Qingyuan Han; Yuhao Lou; Tianyu Zhu; et al. Horticulture Research 2026. doi:10.1093/hr/uhag040

Other claims from this paper (19)

Method / resource PeachDB provides 246 genomic selection models.
Method / resource The citric acid prediction accuracy was 0.74.
Method / resource The malic acid prediction accuracy was 0.71.
Method / resource The quinic acid prediction accuracy was 0.57.
Method / resource SVR had superior predictive performance.
Method / resource DNNGP had superior predictive performance.
Method / resource The citric acid model reached only 0.22 maximum accuracy in the hybrid population.
Method / resource All evaluated malic acid models exceeded 0.7 accuracy in the hybrid population.
Method / resource The highest malic acid model accuracy reached 0.83.
Association Five molecular markers correlated with organic acid content variation.
Observation Fifty-nine of 155 hybrid offspring carried the low-acid genotype at all five markers.
Intervention effect In 2018 GS selected fruit had 16.5% lower mean total acid content than MAS selected fruit.
Intervention effect In 2019 GS selected fruit had 13.13% lower mean total acid content than MAS selected fruit.
Intervention effect Overall GS selected fruit had 14.12% lower mean total acid content than MAS selected fruit.
Method / resource GS screening required 86% fewer low-acid individuals for retention than MAS.
Method / resource GS reduced genotyping cost per hybrid offspring by 50% compared with MAS.
Method / resource GS reduced evaluation land area by 86%.
Method / resource GS improved phenotypic performance by 15%.
Method / resource PeachDB makes genomic resources and downloadable selection models publicly accessible.

Method / resourceobservedoriginal resultmeasured

Claim · model-written summary

DNNGP had superior predictive performance.

Material: DNNGPe0
Outcome: superior predictive performancee0

Not reportedTaxonComparator

Verbatim

e0
“Among these, SVR, LGB, and DNNGP demonstrate superior predictive performance.”
✓ located verbatim in the source · abstract, characters 4708–4785

t0
“peach”
✓ located verbatim in the source · title, characters 85–90

Title

PeachDB: accelerating breeding by reducing cost with AI-powered genomic selection in peach

Qingyuan Han; Yuhao Lou; Tianyu Zhu; et al. Horticulture Research 2026. doi:10.1093/hr/uhag040

Other claims from this paper (19)

Method / resource PeachDB provides 246 genomic selection models.
Method / resource The citric acid prediction accuracy was 0.74.
Method / resource The malic acid prediction accuracy was 0.71.
Method / resource The quinic acid prediction accuracy was 0.57.
Method / resource SVR had superior predictive performance.
Method / resource LGB had superior predictive performance.
Method / resource The citric acid model reached only 0.22 maximum accuracy in the hybrid population.
Method / resource All evaluated malic acid models exceeded 0.7 accuracy in the hybrid population.
Method / resource The highest malic acid model accuracy reached 0.83.
Association Five molecular markers correlated with organic acid content variation.
Observation Fifty-nine of 155 hybrid offspring carried the low-acid genotype at all five markers.
Intervention effect In 2018 GS selected fruit had 16.5% lower mean total acid content than MAS selected fruit.
Intervention effect In 2019 GS selected fruit had 13.13% lower mean total acid content than MAS selected fruit.
Intervention effect Overall GS selected fruit had 14.12% lower mean total acid content than MAS selected fruit.
Method / resource GS screening required 86% fewer low-acid individuals for retention than MAS.
Method / resource GS reduced genotyping cost per hybrid offspring by 50% compared with MAS.
Method / resource GS reduced evaluation land area by 86%.
Method / resource GS improved phenotypic performance by 15%.
Method / resource PeachDB makes genomic resources and downloadable selection models publicly accessible.

Method / resourceobservedoriginal resultnegativemeasured

Claim · model-written summary

Existing carotenoid indices had limited predictive performance.

Taxon: maizeZea mayst0
Outcome: limited predictive performancee0

Not reportedComparatorValue

Verbatim

e0
“Existing carotenoid indices exhibited limited predictive performance for the studied samples”
✓ located verbatim in the source · abstract, characters 557–649

t0
“maize carotenoids”
✓ located verbatim in the source · title, characters 30–47

Title

Non-destructive estimation of maize carotenoids using reflectance-based spectral indices

Attila Nagy; Ahmed Elbeltagi; László Radócz; et al. Frontiers in Plant Science 2026. doi:10.3389/fpls.2026.1699049

Other claims from this paper (5)

Method / resourceobservedoriginal resultpositivemeasured

Claim · model-written summary

Weighting QTL increased predictive performance.

Taxon: WheatTriticum aestivumt0
Gene: QTLe0
Outcome: predictive performancee0
Direction: increasede0
Comparator: weighting QTLe0

Verbatim

e0
“weighting QTL increased predictive performance”
✓ located verbatim in the source · abstract, characters 1181–1227

t0
“Wheat”
✓ located verbatim in the source · title, characters 68–73

Title

Improving Genomic Prediction for Pre‐Harvest Sprouting Tolerance in Wheat by Weighting Large‐Effect Quantitative Trait Loci

Jessica K. Moore; Harish K. Manmathan; Victoria A. Anderson; et al. Crop Science 2017. doi:10.2135/cropsci2016.06.0453

Other claims from this paper (4)

Method / resource Kernel color did not improve model accuracy.
Method / resource Combined marker assisted and genomic selection outperformed other methods.
Method / resource Optimum marker number was 4000 SNPs.
Method / resource Genomic selection model accuracies ranged from 0.49 to 0.62.

Method / resourceobservedoriginal resultpositivemeasured

Claim · model-written summary

GBLUP had the best overall predictive performance, with predictive ability 0.48.

Taxon: Eucalyptus spp.Eucalyptust0
Outcome: predictive abilitye0
Value: 0.48e0

Not reportedComparator

Verbatim

e0
“GBLUP exhibited the best overall performance, with a predictive ability of 0.48 and an R2 of 0.21”
✓ located verbatim in the source · abstract, characters 1287–1384

t0
“Eucalyptus spp.”
✓ located verbatim in the source · title, characters 53–68

Title

Genomic Prediction in a Self-Fertilized Progenies of Eucalyptus spp.

Guilherme Ferreira Melchert; Filipe Manoel Ferreira; Fabiana Rezende Muniz; et al. Plants 2025. doi:10.3390/plants14101422

Other claims from this paper (4)

Method / resource GBLUP had an R2 of 0.21.
Method / resource Bayes LASSO had the lowest mean squared error, 3.72.
Method / resource Other models had mean squared errors from 3.72 to 15.50.
Method / resource GBLUP showed better precision in predicting individual performance and balanced parameter performance.

Observationobservedreview summarybasis n/a

Claim · model-written summary

Machine learning models showed strong corn yield predictive performance.

Taxon: CornZea mays subsp. mayst0
Outcome: strong predictive performancee0

Verbatim

e0
“machine learning models demonstrate strong predictive performance”
✓ located verbatim in the source · abstract, characters 765–830

t0
“Corn”
✓ located verbatim in the source · title, characters 51–55

Title

The Application of Machine Learning Techniques for Corn Yield Prediction and Management: A Systematic Literature Review

Lee Carlo F. Simon Journal of Information Systems Engineering & Management 2025. doi:10.52783/jisem.v10i43s.8532

Other claims from this paper (4)

Method / resourceobservedoriginal resultmeasured

Claim · model-written summary

The predictive models had similar predictive ability.

Outcome: predictive ability performancee0
Direction: similare0
Comparator: different predictive modelse0

Not reportedTaxon

Verbatim

e0
“different predictive models had similar predictive ability performance”
✓ located verbatim in the source · abstract, characters 1398–1468

Title

Genomic Selection with Allele Dosage in Panicum maximum Jacq.

Letícia A de C. Lara; Mateus F Santos; Liana Jank; et al. G3 Genes Genomes Genetics 2019. doi:10.1534/g3.118.200986

Other claims from this paper (3)

Method / resource Tetraploid dosage data gave higher predictive ability than diploid dosage data.
Method / resource The authors provide bioinformatic and modeling guidelines for tetraploid dosage.
Observation Genomic selection may lead to additional gains in recurrent selection of P. maximum.

“predictive performance”

Claim types

Direction of the findings

Over time

Claim type

Research area

Polarity

Statistics