Pores in the leaf epidermis called stomata allow plants to take up carbon dioxide for photosynthesis, but are also pathways for water vapor loss. New image acquisition and analysis methods are allowing high-throughput phenotyping of stomatal patterning, which can be applied to better understand the genetic basis of variation in certain species. However, it takes considerable data and effort to train the models and their ability to accurately detect epidermal structures is constrained by the training data. This issue of context dependency, the inability to perform effectively in novel contexts, is the main hurdle preventing widespread adoption of machine learning in high-throughput phenotyping of intraspecific, interspecific, and environmental variation. Here we show the limited ability of a Mask-RCNN tool trained and successfully applied to Zea mays, to analyze images from a closely related grass called Setaria viridis. We then demonstrate successful retraining of the tool to cope with the novel amounts of diversity presented by this new species. The stomatal complexes in optical tomography images of mature Setaria leaves were accurately identified by comparison to expert raters (R 2 = 0.84). This study highlights the challenge of context dependency for widespread application of machine learning tools for phenotyping plant traits, even in closely related species. At the same time, it also provides a new tool that can be applied to leverage Setaria as a model C4 species, and a roadmap for the translation of a machine learning tool to analyze stomatal patterning in diverse datasets of new plant species.
ORCiD: [https://orcid.org/0000-0003-0655-2343] Increasing weather variability is affecting the overall productivity of agriculture. In this scenario, current crop improvement science is essential to improve productivity while retaining the quality of plant products. There has long been an interest in using process-based modeling to examine the interaction between environment and genotype. The methodological challenges to better predict how various environmental conditions may impact novel genotypes and it has been a fundamental barrier to model parameterization. Thus, a phenotypic campaign was conducted to collect a comprehensive physiological dataset from a panel of 25 genotypes (including both breeder panel and diversity panel) in the summer of 2022. Additionally, an unmanned aerial vehicle (UAV) was used to gather remote sensing data. The red-green-blue (RGB) 3D point cloud, NDVI (Normalize difference vegetation Indices), and LIDAR (Light detection and ranging) were also used to identify the trait variations among the genotypes. The data is being analyzed to explain the physiological and phenotypic trait differences. The outcome of this project would help to develop a genetically informed, realistic soybean model. Finally, it will help breeders and growers in locating high-yielding cultivars for the appropriate geographical areas.
Measurement of key crop physiological traits using high resolution aerial imagery with unmanned aerial systems (UASs) holds enormous potential to increase consistency and accuracy of data collected for field evaluations. Here, we demonstrate temporal corn response to fertility treatments using repeated measurements followed by an area-under-the-curve progression analysis. Radiometrically calibrated multispectral datasets were used to calculate standard vegetative indices as well as to leverage models that approximate leaf area, nitrogen content, chlorophyll content, and canopy uniformity. In addition, digital elevation models can be employed to measure relative canopy heights and spatial variability in the field. Taken together, these digital assessments allow for a researcher to have significant insight into experiment outcomes during the growing season, including the identification of relative yield potential. This approach automates and standardizes the acquisition of key phenotypes that can be used to more efficiently evaluate field trials across multi-location programs.
In recent decades, the field of phenomics has lagged behind the advances in genomics, which have become increasingly high-throughput and low-cost. In comparison, manually collected phenotypes are often time-consuming, labor intensive, and more costly to obtain. The development of high-throughput phenotyping platforms (HTPP) are bridging these gaps and enabling improved spatial and temporal resolution for researchers. We used imagery from unoccupied aerial vehicles (UAV) flown over multiple site years in Saskatchewan and Italy to gather data for crop height, area and volume in a lentil diversity panel. We found high correlations for our UAV-derived traits (height & volume) with our manually collected phenotypes (height & biomass). In addition, the high-throughput nature of the UAV allowed for the collection of time-series data which enabled the modelling of growth curves for volume, height and area, which would be impractical under traditional phenotyping procedures given the large population grown in multiple environments. Principal component analysis and hierarchical clustering revealed differential growth strategies amongst our diverse lentil population across contrasting environments. Our study demonstrates the potential for HTPP to obtain data that traditionally require destructive sampling, e.g., volume as a proxy for vegetative biomass, and improve the temporal quality of phenotype data enabling researchers to take their analysis beyond single time points, e.g., model growth curves. In addition, performing our analysis on data from contrasting environments, i.e., Saskatchewan and Italy, has helped elucidate optimal adaptation with regard to growth strategies in lentils.
Root studies in controlled environments are typically conducted either in rhizotrons, pots, or small scale mesocosm systems, like PVC tubes or root boxes. These systems have two limitations for translating results to crop roots grown in fields. First, the size and shape of containers change the root phenotype when plants are in the mature stage. Second, often only one plant is planted per container without interaction among neighboring plants. Therefore, the root architecture observed in these isolated environments has low predictability for the root architecture in a community setting in fields. To better translate the root traits observed in a controlled environment to field observations, we developed a macro-mesocosm system (5.5 m (W) x 6.7 m (L) x 0.7 m (H)) to mimic the real field soil conditions in a greenhouse. We also installed 64 capacitance soil moisture sensors to monitor the whole macro-mesocosm system at 15.24 cm and 38.10 cm soil depths in real-time. We evaluated the phenotypic spectrum in one common bean (Phaseolus vulgaris. L) genotype, SEQ7, in a time series experiment. We grew SEQ7 for two, six, nine, and twelve weeks under sensor-controlled water-stressed and well-watered irrigation regimes. SEQ7 showed four different root architecture types across developmental stages. These four root architecture types are consistent with previous field observation. This novel macro-mesocosm system will be a great setup to study the field dynamics of the root phenotypic spectrum in a controlled environment.
Stomata are the microscopic pores on plant leaves that open or close to regulate the flux of water from leaves. Guard cells of stomata are known to react to environmental conditions such as light and CO2 in order to optimize CO2 uptake and water loss. Stomatal anatomy (aperture, length, width, etc.) influences leaf-level physiology traits including conductance to water. Stomatal anatomy can be visualized in situ by microscopy, but the difficulty of regulating the atmospheric environment of a microscope stage means that the conditions under which imaging is done are rarely physiologically relevant. Alternatively, portable photosynthesis measuring instruments offer a non-destructive estimate of leaf gas exchange, including stomatal conductance, while the leaf experiences tightly controlled steady-state or dynamic environmental conditions. However, these measurements reflect stomatal characteristics in aggregate on a leaf area basis, which are heavily influenced by the mesophyll as well as epidermal structure and function. Observing the behavior of stomata by microscopy simultaneous to controlling the leaf environment and measuring gas exchange fluxes would allow advances in the understanding of leaf structure-function relationships. To reconcile the microscopic stomatal characteristics with leaf-level gas exchange we have combined laser scanning confocal microscopy and gas exchange instruments to simultaneously observe stomatal characteristics (e.g. stomatal aperture, pore depth, closing speed) and leaf-level traits like photosynthesis, transpiration, and stomatal conductance. Results are presented for the use of this approach on diverse plant species.
Globally, water supply is the major limiting factor for crop productivity. Water use efficiency (WUE) is defined as the ratio of photosynthetic carbon gain relative to water vapor loss from the leaf through the stomata to the atmosphere. Improving WUE would slow crop water use and delay the onset of drought stress when water supply does not meet crop demand. Stomatal density is an important factor that influences plant gas exchange efficiency. We have proof-of-concept that reducing stomatal density in sorghum by ubiquitously expressing a synthetic EPF2core gene can increase WUE without any decrease in photosynthetic carbon gain. However, ubiquitous expression of the synthetic EPF gene has unwanted pleiotropic effects on stem development and seed set. In this study, we test whether tissue-specific promoters can be used to isolate the desired leaf phenotypes without causing unwanted side effects. This provides an important step towards engineering stomatal density to improving WUE and protecting C4 crop yields from drought-induced losses today and in a future, warmer climate.
The maize disease lesion mimic mutants spontaneously form lesions on leaf blades and sheaths that strongly resemble the plant's responses to pathogen infection. Variations in lesion morphology, spatiotemporal distribution, and sensitivity to genetic background and weather make them ideal candidates to develop high throughput and high resolution phenotyping methods for individual plants and their organs in unstructured fields. We present three approaches to imaging lesions at different phenotyping scales and image resolution. Each strategy has distinct advantages and poses unique collection and computational challenges. The first is imaging individual leaves ex situ before sexual maturity using reflected light. The challenge is to identify leaves while the lesions are sufficiently separated for easier segmentation, yet numerous enough for good sample size and mature enough to display the range of lesion developmental stages. This is a moderate throughput, moderate resolution strategy. The second is to image plants using UAVs in situ. The challenges are to fly low enough for good lesion resolution while minimizing extraneous movement and to register individual plants and their leaves during the growing season. This is a high throughput, lower to moderate resolution strategy. The third is to image lesions using after-market lenses on cell phones in situ. The challenges are to capture the same region of the leaves over time without interfering with lesion formation and to mosaic the imagery of highly repetitive surface features into a summary view for registration. This is a low throughput, high resolution strategy.
We investigate the robustness of 3D point-based deep learning for organ segmentation of 3D plant models against varying reconstruction quality of the surface. The reconstruction quality is quantified in two ways: 1) The number of acquisitions for partial 3D scans and 2) the amount of noise. High quality models of real rosebush plants are used to collect point clouds in a controlled simulation environment as a way to degrade surface quality systematically. We show that the well-known 3D point-based neural network PointNet++ is capable of operating effectively on low quality and corrupted data for the task of plant organ segmentation. The results indicate that investing on developing deep learning methods has the potential of advancing applications of automated phenotyping, especially for low-quality 3D point clouds of plants. Keywords: plant phenotyping, organ segmentation, robustness analysis, point-based deep learning (a) (b) Figure 1: A 3D rosebush model from ROSE-X data set: (a) point cloud; (b) triangular mesh model.
Climate change and harsh agricultural practices are increasing the amount of salt and heavy metals in soil, drastically decreasing the amount of arable land while simultaneously lowering crop yields. However, some plants grown in poor soil have adapted diverse mechanisms to cope with harsh environments. It has been hypothesized that the biochemical mechanisms responsible for salt tolerance overlaps with heavy metal tolerance, yet the similarities in these mechanisms are still unknown. Lessons from naturally salt and heavy metal tolerant plants can be applied to crops to increase resilience or be used in phytoremediation efforts. Here, we use the salt and heavy metal tolerant plant Cakile maritima as a model system for phytoremediation by using a large-scale multi-omics approach, combining ionomics, metabolomics, transcriptomics, and genomics, to understand the metabolic responses following NaCl and cadmium stress. We have developed an automated pipeline for tracking salinity, as well as using elemental analysis to monitor intracellular concentrations. We will perform RNA-seq to understand patterns of differential gene expression, gather a list of candidate genes, and use comparative genomics to understand the potential influence of ancient polyploidy on stress tolerance. Combining this with metabolomics will enable a fully integrated understanding of salt stress response and allow us to know if Cakile maritima is predisposed for salt stress or has a rapid stress response. Coupling this with transcriptomics will allow us to identify important pathways and neofunctionalized genes that may be specific for C. maritima stress response and be applied to crop species to enhance resilience.
Field Based High Throughput Phenotyping Enables the Discovery of Loci Linked to Senescence and Grain Filling Period ORCiD: [Alper Adak; 0000-0002-2737-8041] Keywords: Grain filling period, field-based high throughput phenotyping, days to senescence, temporal phenotype. Senescence occurs progressively over time and is variable among different genotypes. To examine the temporal and genetic variation of senescence, 280 maize hybrids and 520 maize recombinant inbred lines (RILs) grown in 2017 and 2018 were investigated. Hybrids were grown in late and optimal planting trials; RILs were grown in irrigated and non-irrigated trials, both based on range-row design with two replications. Two types of Unmanned aerial systems (UAS, also known as UAV or drones) were flown over the germplasm between 14 and 20 times respectively. Temporal senescence of each row-plot in hybrids and RILs was scored visually according to percentile scoring using four to five rectified drone images between ~90 and ~130 days after planting. A mechanistic growth model was fit to each genotype using the temporal senescence scores, resulting in 0.94 and 0.97 R 2 for hybrids and RILs. Days to senescence (DTSE) and grain filling period (GFP) were calculated for each row plot using the developed mechanistic growth model. To predict the genotypic value for each RIL and hybrid, a mixed model with three-way interaction model (Genotype*Flight*Environment) was then run. Correlation was calculated ~0.84 and ~0.88 between grain yield and GFP and DTSE in hybrids. A major quantitative trait locus was also discovered on chromosome 1 (295.5 to 296.8 kb; 15% explained) linked to GFP in RILs. GFP is known to be physiologically important, UAS provided an easily scalable measure which can greatly increase the evaluation of variation in the field.
ORCiD: [0000-0002-7283-3357] Plant Scientists are striving to improve crop response to abiotic stress under adverse environmental conditions. Many bio-physical , biochemical, and physiological traits are difficult to quantify due to the low throughput and destructive nature for their measurements. This study aims to characterize biophysical and physiological traits of maize plants using RGB and hyperspectral imaging in greenhouse condition. Single hybrid Maize genotype with four different treatment combination of water and nitrogen were tested. Plants were imaged, harvested, and measured at several growth stages range from V6 to R5 stages. Images were analyzed and correlation was established between manually measured plant traits and pixel level information extracted from the plants. RGB images are processed to determine projected plant area which are correlated with destructively measured plant shoot fresh weight, dry weight, and biomass area. Hyperspectral images are processed to extract plant leaf reflectance and correlated with leaf nitrogen/chlorophyll content. PLSR models are calibrated to estimate corn leaf nitrogen/chlorophyll content from image-generated hyperspectral data, as well as the leaf hyperspectral data from a handheld ASD spectrometer and their performance will be compared. Biological science, computer vision, mathematics and engineering can be integrated as a holistic approach for quantifying the overall growth, development, and response of maize plants under differential nitrogen rates.
Predictive biology is the ability to predict a biological outcome from known inputs (and vice versa). Complex and urgent problems, such as climate change and a growing global population, requires a better grasp of predictive biology approaches. However, predictive biology requires a deep understanding of the genome-to-phenome relationship within an organism. The field of genomics has accelerated rapidly in the last few decades with technological advances that have helped reduced the costs of genomics research and easy-to-use computational tools. Phenomics technologies has not advanced at the same rate. Many phenotyping systems are expensive to own, require special training to use, focus on narrow areas of research, and produce large amounts of data with no standards of storage and documentation. We propose a design philosophy for high-throughput phenotyping systems called the OPEN Series. This philosophy focuses on systems that use off-the-shelf commercial products and open-source software to make high quality phenotyping systems efficiently and for use by general users. In addition, the OPEN Series focuses on integrating cloud-based image processing through the NSF-funded cyberinfrastructure CyVerse, thus allowing users to share and process data remotely. We've worked to integrate this philosophy into our own phenotyping systems, OPEN leaf and OPEN root, to great success. We hope to export our work in creating accessible and affordable phenotyping system to labs across the globe to accelerate our understanding of the genome-to-phenome relationship for predictive biology.
Environmental stressors play a major role in determining plant phenotypic traits over the growing season. Nitrogen, a critical nutrient for plant growth, imposes serious stresses when applied in insufficient or excessive quantities in crop fields, which can directly reduce yield and the grower's profits. Due to significant variations in developmental rates and nitrogen needs across rice genotypes and environments, field phenotyping can become a challenging task. Non-linear manifold learning approaches were evaluated to account for variation in developmental rate using multispectral, multitemporal imagery acquired by an unmanned aerial vehicle (UAV). Imagery (13-16 flyovers per plot) and ground-truth data (including biomass nitrogen content, leaf chlorophyll content, and yield) were collected on 2,345 rice research plots during the 2021 and 2022 growing seasons. To increase the size of the training set, we also evaluated inclusion of similarly sized tiles sampled from commercial row-rice fields imaged between 2019 and 2022. We compared several supervised machine learning approaches with unsupervised strategies based on manifold learning to evaluate the hypothesis that model architectures which account for variation in developmental trajectories improve prediction of yield variation in response to nitrogen addition. This work will contribute to improved nitrogen rate recommendations using UAV-acquired imagery. Optimized nitrogen prescriptions ensure satisfying the plants' nitrogen needs and contribute to reduced environment impacts and production costs.
From preventing scurvy to being part of religious rituals, citrus are intrinsically connected to human health and perception. From tiny mandarins to head-sized pummelos, citrus capability of hybridization provides a vastly diverse array of fruit sizes and shapes, which in turn corresponds to a diversity of flavors and aromas. These sensory qualities are tightly linked to oil glands in the citrus skin. The oil glands are also key to understanding fruit development, and the essential oils contained by them are fundamental in the food and perfume industries. We study the shape of citrus based on 3D X-ray CT scan reconstruction of 163 different citrus samples comprising 58 different species and cultivars, including samples of all fundamental citrus species. First, using the power of X-rays and image processing, we are able to compare and contrast size ratios between different tissues, such as the size of the skin compared to the rind or the flesh. Second, we model the fruit shape as an ellipsoidal surface, and later we study and infer possible oil gland distributions on this surface using principles of directional statistics. We finally compare and contrast these overall fruit shape models along their gland distributions across different citrus species. This morphological modeling will allow us later to link genotype with phenotype, furthering our insight on how the physical shape is genetically specified in DNA.
Stomatal conductance (SC) was utilized to indicate the rate of gas and water exchange through stomata on the leaf surface. When crops are experiencing water stress during the daytime, their stomata will close to prevent water loss. However, the crop will also receive less CO2 for photosynthesis under this condition. Consequently, accurate and efficient SC prediction can improve irrigation efficiency, particularly in the present day when water is scarce. The common way to estimate SC nowadays is to make predictions using other variables that can be measured quickly, such as conventional regression analysis. The limitation of conventional regression analysis is that it does not account for changes in SC when crops are subjected to both prolonged drought and high temperature stress. We intend to use time series prediction techniques, such as Long Short-Term Memory (LSTM), to determine the correlation between variables at different time scales. The objective of this study is to (1) Investigating the relationship between SC and persistent weather patterns. (2) Predict SC based on continuously collected soil moisture, plant canopy temperature and weather information. In this study, we measured the SC of soybean and maize by using a handheld leaf porometer. This was then compared to short-term historical crop SC predictions based on canopy temperature, soil moisture, and weather conditions. Autoregressive Integrated Moving Average (ARIMA) and LSTM based on Recurrent neural network (RNN) are used to predict SC, and the results are compared with the conventional modeling method. More details will be presented at the NAPPN conference.