Enhancing prediction accuracy of key biomass partitioning traits in wheat using multi-kernel genomic prediction models integrating secondary traits and environmental covariates

Authors: KUNWAR, SUDIP - University Of Florida; ALI BABAR, MD - University Of Florida; JARQUIN, DIEGO - University Of Florida; AMPATZIDIS, YIANNIS - University Of Florida; KHAN, NAEEM - University Of Florida; PRABHAT ACHARYA, JANAM - University Of Florida; MCBREEN, JORDAN - University Of Florida; ADEWALE, SAMUEL - University Of Florida; Brown Guedira, Gina

Submitted to: The Plant Genome
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 4/9/2025
Publication Date: 6/3/2025
Citation: Kunwar, S., Ali Babar, M., Jarquin, D., Ampatzidis, Y., Khan, N., Prabhat Acharya, J., Mcbreen, J., Adewale, S., Brown Guedira, G.L. 2025. Enhancing prediction accuracy of key biomass partitioning traits in wheat using multi-kernel genomic prediction models integrating secondary traits and environmental covariates. The Plant Genome. https://doi.org/10.1002/tpg2.70052.
DOI: https://doi.org/10.1002/tpg2.70052

Interpretive Summary: Improving the way wheat plants distribute resources to different parts can lead to higher grain yields. The spike partitioning index (SPI) measures how resources are distributed to the spike during flowering, which is crucial for the survival of florets and the formation of grains. Harvest index (HI) and fruiting efficiency (FE) evaluate how well the resources available at the spike during flowering are converted into grains. In this study, we tested 341 wheat lines over three years in Florida to predict these traits. Advanced models combining genomic data, environmental covariates, and secondary traits achieved up to 78% accuracy, compared to 43% or less with simpler genomics-only models. These comprehensive models effectively ranked wheat lines for the different partitioning traits. This research showed that integrating diverse data types or omics helps to predict complex partitioning traits in wheat and aids breeders in decision-making, paving the way for higher-yielding wheat crops in the future.

Technical Abstract: Achieving significant genetic gains in grain yield (GY) in wheat (Triticum aestivum L.) requires optimization of the key biomass partitioning traits such as spike partitioning index (SPI) and fruiting efficiency (FE). However, traditional manual phenotyping of these traits is labor-intensive and destructive, making it unsuitable for evaluating large germplasm panels. This study developed genomic prediction models to estimate these traits using diverse statistical methods while enhancing predictive ability (PA) by integrating environmental covariates (ECs) and secondary traits. A panel of 341 soft wheat elite lines was evaluated for biomass partitioning and yield-related traits from 2022 to 2024 in Citra, Florida. Genomic best linear unbiased predictor (GBLUP) and Bayesian methods performed similarly or better than machine learning models for spike partitioning index (SPI), harvest index (HI), and grain yield (GY). On the other hand, random forest models performed better in predicting effective tillers m-2 (ET), thousand-grain weight (TGW), and grains m-2 (GN). Genomics-only models showed moderate predictive ability (PA), but integrating ECs and secondary traits (i.e. plant height-PH and above-ground-biomass-AGDM) in multi-kernel models improved PA significantly, reaching up to 78%, a substantial improvement over the 43% achieved with genomic-only models. Validations performed on independent datasets confirmed the reliability of the multi-kernel models, even though they showed a slightly lower predictive accuracy compared to within-panel validations. These findings highlight the potential of integrating diverse data types or omics to enhance the prediction of biomass partitioning traits, speeding up genetic advancements, and the development of high-yield wheat varieties to address future food security challenges.