High-throughput phenotyping of stay-green in a sorghum breeding program using unmanned aerial vehicles and machine learning

Authors: Pugh, Nicholas; Young, Andrew; Emendack, Yves; Sanchez, Jacobo; Xin, Zhanguo; Hayes, Chad

Submitted to: The Plant Phenome Journal
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 11/28/2024
Publication Date: 1/13/2025
Citation: Pugh, N.A., Young, A.W., Emendack, Y., Sanchez, J., Xin, Z., Hayes, C.M. 2025. High-throughput phenotyping of stay-green in a sorghum breeding program using unmanned aerial vehicles and machine learning. The Plant Phenome Journal. 8(1). https://doi.org/10.1002/ppj2.70014.
DOI: https://doi.org/10.1002/ppj2.70014

Interpretive Summary: Sorghum is a critical feed and forage crop for the future, particularly as the climate becomes warmer and drier. In particular, "stay-green", or late season drought tolerance, is a critical trait for sorghum breeders and producers. However, most methods to evaluate stay-green are based on subjective rating scales or are time consuming. The use of unmanned aerial vehicles (UAV), or "drones", to rapidly collect information on a field, can help to increase the efficiency of stay-green evaluation in breeding programs. In addition, the use of artificial intelligence models that can make predictions of stay-green using input data could allow breeders and researchers to estimate stay-green without the need to physically enter a field. We find that the two models we tested in this study, random forest and XGBoost, were able to estimate stay-green to a high degree of precision and accuracy. Although the models show great promise for aiding in stay-green measurements, it is important that we continue to update the models as more data is collected so that the model is able to handle a wide range of circumstances. This research will help sorghum breeders produce higher yielding and more drought tolerant sorghum.

Technical Abstract: As climate change continues to influence global weather patterns, the frequency and severity of drought conditions are expected to increase, posing a significant challenge to crop production. In sorghum, a key cereal crop, the stay-green trait is of particular importance as a measure of how well a genotype can tolerate post-anthesis drought conditions, which are critical for final yield. Despite its importance, there is a pressing need for a more efficient, accurate, and precise method to phenotype stay-green in sorghum to enhance breeding efforts. To address this need, this study explores the application of random forest and XGBoost machine learning models for phenotyping the stay-green trait in sorghum. These models provide quantitative measurements that have the potential to enhance genomic studies and offer additional benefits. Although correlations with vegetation indices were occasionally high, they were not sufficiently reliable to be used exclusively. The machine learning models, in contrast, showed high percentages of genetic variation explained and had high repeatability. The values generated by these algorithms enable plant breeders to efficiently make selections in their stay-green breeding programs. Further research is needed to assess the robustness of these models across different environments and genetic material. Additionally, comparing these models with other machine learning approaches will help determine if decision tree-based models are the most effective for this application. Overall, the models presented in this study serve as a promising foundation for improving the efficiency of stay-green breeding programs in sorghum, but they require further validation and comparison with alternative approaches.