Location: Plant Stress and Germplasm Development Research
2025 Annual Report
Objectives
Objective 1. Use traditional and molecular breeding techniques to develop superior sorghum breeding germplasm and hybrids adapted to diverse environments including improved thermal tolerance and improved water use efficiency.
Sub-objective 1A: Development and agronomic testing of grain and forage inbreds and hybrids with the dominant multiple tiller trait.
Objective 2. Utilize diverse germplasm and sorghum mutants to discover and characterize genes and traits such as cold and drought tolerance, and improved hybrid yield, required for superior sorghum production.
Sub-objective 2A: Characterize phenology and biomass accumulation of a dominant multiple tiller mutant (mtl-d1).
Sub-objective 2B: Using remote sensing via small unoccupied aircraft systems (sUAS) as a high-throughput method to screen stay-green and sugarcane aphid (SCA) tolerance.
Sub-objective 2C: Identify the causal mutation for mtl-d1 and genes associated with biomass production.
Sub-objective 2D: Evaluation and analyses of morphological variation in grain composition and total seed protein and quality for 256 sequenced AIMS sorghum mutants.
Sub-objective 2E: Identification of candidate genes and development of DNA markers for increased seed protein content in sorghum.
Objective 3. Discover and characterize sorghum physiological adaptation traits such as modified leaf angle, variable stomatal density, and stay-green drought tolerance, in diverse sorghum germplasm.
Sub-objective 3.A: Characterizing radiation use efficiency (RUE) in sorghum through erect leaf architecture and plant height.
Sub-objective 3B: Exploring potential increase in sink size and strength by characterizing components of panicle architecture using automated tools.
Approach
Sorghum (Sorghum bicolor, L. Moench) is an important C4 crop that is grown in a variety of environments worldwide for food, feed, forage, and cellulosic biomass production. The crop is known for its inherent drought tolerance, especially compared to other cereal crops like maize and rice. More recently, sorghum has gained attention as a health food crop with desirable market characteristics such as gluten-free grain, and as a sustainably grown product that is rich with beneficial compounds such as antioxidants. Sorghum, as a crop commodity in the United States, is of critical importance in major grain production regions of the country where water is limited. Unfortunately, sorghum crop improvement has remained relatively stagnant for the last 40 years. This stagnation in yield improvements is partially due to minimal research investments by public and private institutions. As climate change effects agricultural production worldwide, and specifically the U.S. Great Plains, it is imperative that scientists improve sorghum in terms of yield potential and end-user utilization. The objective of this research is to integrate recent advances in plant breeding and molecular biology with next generation phenotyping technologies to accelerate the rate of genetic gain in grain and forage sorghum. This project aims to elucidate the genes and gene networks controlling novel agronomic and compositional traits such as leaf erectness, altered and optimized plant height, multi-tillering, and grain protein enhancement. The products of this research will include improved sorghum germplasm, trait-specific genetic markers, and an improved understanding of the physiological traits that enhance sorghum productivity.
Progress Report
Objective 1: ARS scientists in Lubbock, Texas, continue testing experimental sorghum hybrids containing the multiple tiller trait. Advanced hybrid evaluation of diverse sorghum forage hybrids is being performed across four locations in Texas. Hybrids are being evaluated for confirmation of trait expression in multiple environments, response to variable day length conditions, and for forage yield and quality. Preliminary results indicate that the trait is expressed in many different growing environments, and that there is an increase in tillers and above-ground biomass, but more research is needed to confirm forage yield and quality. Studies focused on the agronomic performance of multiple tiller sorghum hybrids are essential for identifying the benefits of the trait, especially considering the delayed anthesis and prolonged crop maturity consistently observed with the trait.
Objective 2: Sub-Objective 2A: ARS scientists in Lubbock, Texas, continue testing above-ground and below-ground biomass of the multiple tiller trait in diverse environments. Researchers have confirmed that the multiple tiller trait produces an increase in both leaf and root biomass in forage, and grain sorghum genetic backgrounds. Researchers validated previous observations that anthesis is delayed approximately 18-21 days compared to the wild-type. Researchers continue to evaluate the potential agronomic consequences of delayed anthesis and are also evaluating if the anthesis timing can be shortened through breeding. Preliminary results from six diverse breeding populations indicate delayed anthesis is indicative of the trait, and that breeding for early maturity may not be possible.
Objective 2: Subobjective 2B: ARS scientists in Lubbock, Texas, continue to utilize drones to accurately screen diverse sorghum germplasm for post-flowering drought tolerance. The novel methods developed by researchers in Lubbock, Texas, have demonstrated that over 20,000 sorghum plots can be quickly and accurately screened for post-flowering drought tolerance without the need of time-consuming visual rating systems. Multi-environment results continue to confirm that unmanned aerial vehicle screening for drought tolerance is very similar to ground-truth data in terms of accuracy. The recent acquisition of a multi-spectrum UAV to accompany our visual light spectrum UAV is anticipated to further refine and improve prediction accuracies.
Objective 2: Subobjective 2D and 2E: ARS scientists in Lubbock, Texas, grew-out and collected grain for 256 fully sequenced sorghum mutants. Each mutant line was analyzed by near-infrared spectroscopy for important grain quality traits such as protein and starch composition. This vital data will allow researchers to investigate the genetic control of important grain composition traits in sorghum. For Sub-objective 2E, ARS scientists in Lubbock, Texas, initiated the development of genetic mapping populations for identifying genes controlling grain protein in sorghum. Initial backcrossing to the wild-type parent has been completed.
Objective 2: Subobjective 3A and 3B: Leaf angles of 50 lines with varying degrees of leaf angles (1st leaf below flag leaf) and internode length between 3rd and 4th (2nd and 3rd leaves respectively below flag leaf) leaves was measured. A LI-600 porometer with GPS accelerometer/magnetometer was used to calculate leaf’s angle of incidence to sun, leaf pitch, roll, heading, slope, and the effects on stomata conductance, transpiration, photosystem II, electron transport rate, lead vapor pressure deficit, and leaf temperature. Additionally, Images from 1264 panicles from 316 genotypes from multi-location trials were generated for fully automated deep learning tool for sorghum grain count. Manual annotated images were used to train and generate a model for sorghum panicle grain count.
Accomplishments
1. Advancing sorghum grain count estimation. Accurately estimating panicle grain count—a trait that directly drives overall yield in sorghum—has long been limited by the absence of robust benchmark image datasets. To address this, ARS researchers in Lubbock, Texas, in collaboration with Kansas State University, Texas Tech University, and Texas A&M University, developed the Sorghum Grain Count (SGC) dataset: a comprehensive image collection featuring both front and back views of 1,264 panicles from 316 distinct genotypes. These results pave the way for reliable and scalable grain count estimation—not only for sorghum, but potentially for other crops as well. By combining this landmark dataset with the advanced methods, researchers aim to propel machine learning and computer vision to the forefront of agricultural science. This work offers a transformative approach to yield prediction and marks a critical step in the global movement toward precision agriculture.
2. Commercialization of a high yielding grain sorghum hybrid with sorghum-aphid resistance. ARS scientists in Lubbock, Texas, have developed, and licensed a high-yielding grain sorghum hybrid to industry partners. The development of a high-yielding, early maturing grain sorghum hybrid with sorghum aphid resistance represents a critical advancement in sorghum genetic improvement, providing tangible benefits for growers, and stakeholders.
3. Development and evaluation of machine learning models to estimate drought tolerance in sorghum. To streamline plant breeding pipelines and empower plant breeders with data-driven management decisions, it's essential to develop rapid, non-invasive techniques for estimating key crop traits. One promising approach involves leveraging drone technology—particularly imagery captured across multiple growth stages—to efficiently collect high-quality crop data. ARS scientists in Lubbock, Texas, have optimized the use of drone-collected data to generate accurate and high-throughput field data, offering valuable insight into crop development during drought stress. This methodology produces improved sorghum germplasm for breeding programs to incorporate into new sorghum inbreds and hybrids.
Review Publications
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.
Emendack, Y., Sanchez, J.V., Laza, H.E. 2025. Dhurrin: An endogenous turnover n-source for early seedling growth in sorghum. Frontiers in Plant Science. 16. https://doi.org/10.3389/fpls.2025.1558712.
Goebel, T.S., Mahan, J.R., Payton, P., Young, A.W., Pugh, N.A., Xin, Z., Stout, J.E., Gitz, D.C., Lascano, R.J. 2025. Quantifying temporal distortions of artificial UAV crop canopy temperature measurements. Remote Sensing. 14(2). https://doi.org/10.4236/ars.2025.142006.
Davis, R.F., Harris-Shultz, K.R., Hayes, C.M., Xin, Z., Knoll, J.E. 2025. Evaluating diverse sorghum genotypes used in breeding programs for resistance to Meloidogyne incognita. Nematropica. 54:166-176.
Wang, B., Jiao, Y., Olson, A., Huang, J., Liaca, V., Fengler, K., Wei, X., Wang, L., Wang, X., Regulski, M., Drenkow, J., Gingeras, T., Hayes, C.M., Armstrong, J.S., Huang, Y., Xin, Z., Ware, D., Kumar, V., Chougule, K. 2024. High-quality chromosome scale assemblies of two important sorghum inbred lines Tx2783 and RTx436. The Plant Journal. 6(3). https://doi.org/10.1093/nargab/lqae097.
Tian, R., Najera-Gonzalez, H., Nigam, D., Khan, A., Herrera-Estrella, L., Chen, J., Xin, Z., Jiao, Y. 2024. A Leucine-rich repeat receptor kinase as a regulator in the cuticular wax deposition in sorghum. Journal of Experimental Botany. 75(20):6331-6345. https://doi.org/10.1093/jxb/erae319.