Optimizing genomic prediction with transfer learning under a ridge regression framework

Genomic selection (GS) is a predictive plant and animal methodology that allows the selection of plants and animals based on predictions without the need to measure the phenotype. However, its practical application requires challenging prediction accuracy due to the noise observations collected in e...

Descripción completa

Detalles Bibliográficos
Autores principales: Montesinos-Lopez, Osval A., Barajas-Ramirez, Eduardo A., Salinas Ruiz, Josafhat, Montesinos-López, Abelardo, Gerard, Guillermo S., Vitale, Paolo, Dreisigacker, Susanne, Saint Pierre, Carolina, Crossa, José
Formato: Journal Article
Lenguaje:Inglés
Publicado: Wiley 2025
Materias:
Acceso en línea:https://hdl.handle.net/10568/176260
_version_ 1855533734078447616
author Montesinos-Lopez, Osval A.
Barajas-Ramirez, Eduardo A.
Salinas Ruiz, Josafhat
Montesinos-López, Abelardo
Gerard, Guillermo S.
Vitale, Paolo
Dreisigacker, Susanne
Saint Pierre, Carolina
Crossa, José
author_browse Barajas-Ramirez, Eduardo A.
Crossa, José
Dreisigacker, Susanne
Gerard, Guillermo S.
Montesinos-Lopez, Osval A.
Montesinos-López, Abelardo
Saint Pierre, Carolina
Salinas Ruiz, Josafhat
Vitale, Paolo
author_facet Montesinos-Lopez, Osval A.
Barajas-Ramirez, Eduardo A.
Salinas Ruiz, Josafhat
Montesinos-López, Abelardo
Gerard, Guillermo S.
Vitale, Paolo
Dreisigacker, Susanne
Saint Pierre, Carolina
Crossa, José
author_sort Montesinos-Lopez, Osval A.
collection Repository of Agricultural Research Outputs (CGSpace)
description Genomic selection (GS) is a predictive plant and animal methodology that allows the selection of plants and animals based on predictions without the need to measure the phenotype. However, its practical application requires challenging prediction accuracy due to the noise observations collected in experiments in these areas. Many strategies and approaches have been proposed to improve the prediction accuracy of this methodology. This paper explores the use of transfer learning in the context of GS. Transfer learning with (1) ridge regression (RR) (Transfer RR) and (2) analytic RR (ARR) (Transfer ARR) were applied from cultivars in the proxy environment to predict those cultivars in the goal environments. Also, we compared the performance of models RR and ARR without transfer learning. We used 11 real multi-environment datasets (wheat and rice) and evaluated them in terms of Pearson's correlation (Cor) and normalized root mean square error (NRMSE). Our study shows empirical evidence that the Transfer RR or Transfer ARR approaches significantly enhanced predictive performance. Across the datasets, Transfer RR (or Transfer ARR) method improved Cor by 22.962% and NRMSE by 5.757%, in comparison to models RR and ARR. These results underscore the potential of Transfer RR (or Transfer ARR) when enhancing predictive accuracy in this context.
format Journal Article
id CGSpace176260
institution CGIAR Consortium
language Inglés
publishDate 2025
publishDateRange 2025
publishDateSort 2025
publisher Wiley
publisherStr Wiley
record_format dspace
spelling CGSpace1762602025-12-08T10:11:39Z Optimizing genomic prediction with transfer learning under a ridge regression framework Montesinos-Lopez, Osval A. Barajas-Ramirez, Eduardo A. Salinas Ruiz, Josafhat Montesinos-López, Abelardo Gerard, Guillermo S. Vitale, Paolo Dreisigacker, Susanne Saint Pierre, Carolina Crossa, José genomics forecasting marker-assisted selection learning datasets Genomic selection (GS) is a predictive plant and animal methodology that allows the selection of plants and animals based on predictions without the need to measure the phenotype. However, its practical application requires challenging prediction accuracy due to the noise observations collected in experiments in these areas. Many strategies and approaches have been proposed to improve the prediction accuracy of this methodology. This paper explores the use of transfer learning in the context of GS. Transfer learning with (1) ridge regression (RR) (Transfer RR) and (2) analytic RR (ARR) (Transfer ARR) were applied from cultivars in the proxy environment to predict those cultivars in the goal environments. Also, we compared the performance of models RR and ARR without transfer learning. We used 11 real multi-environment datasets (wheat and rice) and evaluated them in terms of Pearson's correlation (Cor) and normalized root mean square error (NRMSE). Our study shows empirical evidence that the Transfer RR or Transfer ARR approaches significantly enhanced predictive performance. Across the datasets, Transfer RR (or Transfer ARR) method improved Cor by 22.962% and NRMSE by 5.757%, in comparison to models RR and ARR. These results underscore the potential of Transfer RR (or Transfer ARR) when enhancing predictive accuracy in this context. 2025-09 2025-08-29T16:35:43Z 2025-08-29T16:35:43Z Journal Article https://hdl.handle.net/10568/176260 en Open Access application/pdf Wiley Montesinos‐López, O. A., Barajas‐Ramirez, E. A., Salinas‐Ruiz, J., Montesinos‐López, A., Gerard, G., Vitale, P., Dreisigacker, S., Pierre, C. S., & Crossa, J. (2025). Optimizing genomic prediction with transfer learning under a ridge regression framework. The Plant Genome, 18(3). https://doi.org/10.1002/tpg2.70049
spellingShingle genomics
forecasting
marker-assisted selection
learning
datasets
Montesinos-Lopez, Osval A.
Barajas-Ramirez, Eduardo A.
Salinas Ruiz, Josafhat
Montesinos-López, Abelardo
Gerard, Guillermo S.
Vitale, Paolo
Dreisigacker, Susanne
Saint Pierre, Carolina
Crossa, José
Optimizing genomic prediction with transfer learning under a ridge regression framework
title Optimizing genomic prediction with transfer learning under a ridge regression framework
title_full Optimizing genomic prediction with transfer learning under a ridge regression framework
title_fullStr Optimizing genomic prediction with transfer learning under a ridge regression framework
title_full_unstemmed Optimizing genomic prediction with transfer learning under a ridge regression framework
title_short Optimizing genomic prediction with transfer learning under a ridge regression framework
title_sort optimizing genomic prediction with transfer learning under a ridge regression framework
topic genomics
forecasting
marker-assisted selection
learning
datasets
url https://hdl.handle.net/10568/176260
work_keys_str_mv AT montesinoslopezosvala optimizinggenomicpredictionwithtransferlearningunderaridgeregressionframework
AT barajasramirezeduardoa optimizinggenomicpredictionwithtransferlearningunderaridgeregressionframework
AT salinasruizjosafhat optimizinggenomicpredictionwithtransferlearningunderaridgeregressionframework
AT montesinoslopezabelardo optimizinggenomicpredictionwithtransferlearningunderaridgeregressionframework
AT gerardguillermos optimizinggenomicpredictionwithtransferlearningunderaridgeregressionframework
AT vitalepaolo optimizinggenomicpredictionwithtransferlearningunderaridgeregressionframework
AT dreisigackersusanne optimizinggenomicpredictionwithtransferlearningunderaridgeregressionframework
AT saintpierrecarolina optimizinggenomicpredictionwithtransferlearningunderaridgeregressionframework
AT crossajose optimizinggenomicpredictionwithtransferlearningunderaridgeregressionframework