Machine Learning-Based identification of resistance genes associated with sunflower broomrape DOI Creative Commons

Yingxue Che,

Congzi Zhang,

Jixiang Xing

et al.

Plant Methods, Journal Year: 2025, Volume and Issue: 21(1)

Published: May 16, 2025

Sunflowers (Helianthus annuus L.), a vital oil crop, are facing severe challenge from broomrape (Orobanche cumana), parasitic plant that seriously jeopardizes the growth and development of sunflowers, limits global production leads to substantial economic losses, which urges resistant sunflower varieties. This study aims identify resistance genes comprehensive transcriptomic profile 103 varieties based on gene expression data then constructs predictive models with key genes. The least absolute shrinkage selection operator (LASSO) regression random forest feature importance ranking method were used These considered as biomarkers in constructing machine learning Support Vector Machine (SVM), K-Nearest Neighbours (KNN), Logistic Regression (LR), Gaussian Naive Bayes (GaussianNB). SVM model constructed 24 selected by LASSO demonstrated high classification accuracy (0.9514) robust AUC value (0.9865), effectively distinguishing between susceptible data. Furthermore, we discovered correlation differential metabolites, particularly jasmonic acid (JA). Our highlights novel perspective screening for resistance, is anticipated guide future biological research breeding strategies.

Language: Английский

MtCro: multi-task deep learning framework improves multi-trait genomic prediction of crops DOI Creative Commons

Dian Chao,

Hao Wang,

Fengqiang Wan

et al.

Plant Methods, Journal Year: 2025, Volume and Issue: 21(1)

Published: Feb. 5, 2025

Genomic Selection (GS) predicts traits using genome-wide markers, speeding up genetic progress and enhancing breeding efficiency. Recent emphasis has been placed on deep learning models to enhance prediction accuracy. However, current focus specific phenotypes for the given task, overlooking inter-correlations among different phenotypes. In response, we introduce MtCro, a multi-task approach that simultaneously captures diverse plant within shared parameter space. Extensive experiments reveal MtCro outperforms mainstream models, including DNNGP SoyDNGP, with performance gains of 1-9% Wheat2000 dataset, 1-8% Wheat599, 1-3% Maize8652. Furthermore, comparative analysis shows consistent 2-3% improvement in multi-phenotype predictions, emphasizing impact inter-phenotype correlations By leveraging learning, efficiently phenotypes, both model training efficiency accuracy, ultimately accelerating breeding. Our code is available https://github.com/chaodian12/mtcro .

Language: Английский

Citations

1
Advances in Machine Learning for Epigenetics and Biomedical Applications DOI
Hao Lin, Hao Lv, Fanny Dao

et al.

Methods, Journal Year: 2025, Volume and Issue: 235, P. 53 - 54

Published: Feb. 1, 2025

Language: Английский

Citations

0
PhytoCluster: a generative deep learning model for clustering plant single-cell RNA-seq data DOI Creative Commons
Haolin Wang, Xiangzheng Fu, Lijia Liu

et al.

aBIOTECH, Journal Year: 2025, Volume and Issue: unknown

Published: Feb. 20, 2025

Language: Английский

Citations

0
Machine Learning-Based identification of resistance genes associated with sunflower broomrape DOI Creative Commons

Yingxue Che,

Congzi Zhang,

Jixiang Xing

et al.

Plant Methods, Journal Year: 2025, Volume and Issue: 21(1)

Published: May 16, 2025

Sunflowers (Helianthus annuus L.), a vital oil crop, are facing severe challenge from broomrape (Orobanche cumana), parasitic plant that seriously jeopardizes the growth and development of sunflowers, limits global production leads to substantial economic losses, which urges resistant sunflower varieties. This study aims identify resistance genes comprehensive transcriptomic profile 103 varieties based on gene expression data then constructs predictive models with key genes. The least absolute shrinkage selection operator (LASSO) regression random forest feature importance ranking method were used These considered as biomarkers in constructing machine learning Support Vector Machine (SVM), K-Nearest Neighbours (KNN), Logistic Regression (LR), Gaussian Naive Bayes (GaussianNB). SVM model constructed 24 selected by LASSO demonstrated high classification accuracy (0.9514) robust AUC value (0.9865), effectively distinguishing between susceptible data. Furthermore, we discovered correlation differential metabolites, particularly jasmonic acid (JA). Our highlights novel perspective screening for resistance, is anticipated guide future biological research breeding strategies.

Language: Английский

Citations

0