Search for: All records

Abstract Repairing a damaged body part is critical for the survival of any organism. In plants, tissue damage induces rapid responses that activate defense, regeneration and wound healing. While early wound signaling mediated by phytohormones, electrical signals and reactive oxygen species is well-characterized, the mechanisms governing the final stages of wound healing remain poorly understood. Here, we show that wounding in Arabidopsis leaves induces localized cooling, likely due to evaporative water loss, accompanied by the activation of cold-responsive genes. The subsequent disappearance of localized cooling and deactivation of cold-responsive genes serve as a quantitative marker of wound healing. Based on these observations, we developed a workflow by leveraging computer vision and deep learning to monitor the dynamics of wound healing. We found that CBFs transcription factors relay injury-induced cooling signal to wound healing. Thus, our work advances our understanding of tissue repair and provides a tool to quantify wound healing in plants. 

Abstract BackgroundPlant architecture can influence crop yield and quality. Manual extraction of architectural traits is, however, time-consuming, tedious, and error prone. The trait estimation from 3D data addresses occlusion issues with the availability of depth information while deep learning approaches enable learning features without manual design. The goal of this study was to develop a data processing workflow by leveraging 3D deep learning models and a novel 3D data annotation tool to segment cotton plant parts and derive important architectural traits. ResultsThe Point Voxel Convolutional Neural Network (PVCNN) combining both point- and voxel-based representations of 3D data shows less time consumption and better segmentation performance than point-based networks. Results indicate that the best mIoU (89.12%) and accuracy (96.19%) with average inference time of 0.88 s were achieved through PVCNN, compared to Pointnet and Pointnet++. On the seven derived architectural traits from segmented parts, an R²value of more than 0.8 and mean absolute percentage error of less than 10% were attained. ConclusionThis plant part segmentation method based on 3D deep learning enables effective and efficient architectural trait measurement from point clouds, which could be useful to advance plant breeding programs and characterization of in-season developmental traits. The plant part segmentation code is available athttps://github.com/UGA-BSAIL/plant_3d_deep_learning. 

Abstract Yield improvement in cotton could be accelerated through selection for functional yield drivers such as interception of cumulative photosynthetically active radiation (∑IPAR), radiation use efficiency (RUE), and harvest index (HI). However, information on the extent to which these traits vary in cotton in the southeastern United States is limited. It was hypothesized that functional yield drivers would vary significantly within a diverse cotton collection. This study was conducted in Tifton and Athens, GA, and included a total of 4 site‐years. Lint yield, total biomass production, ∑IPAR, RUE, and HI were all affected by genotype. Biomass was more strongly correlated with RUE than ∑IPAR. Even among the highest yielding genotypes, values for functional yield drivers (biomass and harvest index) differed significantly, indicating that high yields could be achieved by differentially manipulating these underlying traits. However, when considered for all genotypes, only HI exhibited a significant positive correlation with yield. Boll production and intra‐boll yield components were also affected by genotype. When considered across upland genotypes, lint per boll, lint per seed, and lint percent were strongly associated with HI and lint yield, whereas boll mass and seed number per boll were not. We conclude that the genotypes evaluated in the current study achieve high lint production per boll and lint yields by manipulating different yield drivers. However, lint yield was primarily maximized through an increase in HI due to increases in boll production and within‐boll distribution of biomass to fiber, not due to increases in total biomass production or boll size.