The value of training population optimization strategies for genomic prediction in the era of large integrated data
The Crop Journal · 2026-09
DOI: 10.1016/j.cj.2026.08.012 ↗Kurz gefasst
Fachartikel in der Zeitschrift The Crop Journal, 2026-09. Verfasst von Ravindra Reddy Gundala und 2 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 79 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0). Gefördert von: Federal Ministry of Education and Research Bonn Office, Federal Ministry of Food and Agriculture, Bundesministerium für Ernährung und Landwirtschaft.
Abstract (Originalsprache)
This study aimed to identify a training population optimization (TPO) strategy that maximizes genomic prediction ability using an independent, representative elite test set and a large integrated training set. Applied to sampled large integrated datasets, each comprising 85% of a large integrated wheat dataset of 18,000 genotypes evaluated for plant height and grain yield across multiple environments, we compared 14 TPO methods to predict the performance of representative German elite genotypes tested in multiple environments across Germany. The optimal TPO strategy differed between traits, but optimized training populations achieved higher prediction ability than the sampled large integrated dataset while using substantially fewer genotypes. For grain yield and plant height, optimized training populations comprising only 32% and 52% of the available genotypes increased prediction ability by 6% and 4%, respectively. These results demonstrate that carefully selected training populations can achieve more accurate genomic predictions with substantially fewer genotypes than large integrated datasets. Based on these results, we provide practical recommendations for selecting suitable TPO methods under various breeding scenarios.
Förderung
- Federal Ministry of Education and Research Bonn Office
- Federal Ministry of Food and Agriculture
- Bundesministerium für Ernährung und Landwirtschaft
Zitieren
APA
Gundala, R. R., Reif, J. C., & Zhao, Y. (2026). The value of training population optimization strategies for genomic prediction in the era of large integrated data. The Crop Journal. https://doi.org/10.1016/j.cj.2026.08.012
Vancouver
Gundala RR, Reif JC, Zhao Y. The value of training population optimization strategies for genomic prediction in the era of large integrated data. The Crop Journal. 2026. doi:10.1016/j.cj.2026.08.012
BibTeX
@article{Gundala2026,
author = {Gundala, Ravindra Reddy and Reif, Jochen Christoph and Zhao, Yusheng},
title = {The value of training population optimization strategies for genomic prediction in the era of large integrated data},
journal = {The Crop Journal},
year = {2026},
doi = {10.1016/j.cj.2026.08.012}
}
EUVIMED