Plant growth and development is impacted by the ability to capture resources including sunlight, determined in part by the arrangement of plant parts throughout the canopy. This is a very complex trait to describe, but has a major impact on downstream traits such as biomass or grain yield per acre. Though some is known about genetic factors contributing to leaf angle, maturity, and leaf size and number, these discrete traits do not encompass the structural complexity of the canopy. In addition, modeling and prediction for plant developmental traits using genomics or phenomics are usually conducted separately. We have developed proof-of-concept models that incorporate spatio-temporal factors from drone-acquired LiDAR features in a maize diversity panel to predict plant growth and development over time to improve our understanding of the biology of canopy formation and development. Briefly, voxel models for probability of beam penetration into the foliage were generated from 3D LiDAR scans collected at seven dates throughout crop canopy development. From the same plots, key architectural features of the maize canopy were measured by hand: stand count; plant, tassel, and flag leaf height; anthesis and silking dates; ear leaf, total leaf, and largest leaf number; and largest leaf length and width. We develop a self-supervised autoencoding neural network architecture that separately encodes plant temporal growth patterns for individual genotypes and plant spatial distributions for each plot. Then, leveraging the resulting latent space encoding of the LiDAR scans, we train and demonstrate accurate prediction of hand-measured crop traits.

Roots are the interface between the plant and the soil and play a central role in multiple ecosystem processes. With intensification of agricultural practices, rhizosphere processes are being disrupted and are causing degradation of the physical, chemical, and biotic properties of soil. Improvement of ecosystem service performance is rarely considered as a breeding trait due to the complexities and challenges of belowground evaluation. Advancements in root phenotyping and genetic tools are critical in accelerating ecosystem service improvement in cover crops. Here I will present root phenotyping approaches for assessing ecosystem service in a prospective cash cover crop; pennycress (Thlaspi arvense L.). In development is a large format mesocosm system that will allow 3D root system architecture analysis of multiple plants. Using this system, we will be assessing how variation in pennycress root system architecture can affect ecosystem service and abiotic stress tolerance with the plant to scale from single plant to canopy level traits.

Tiny tomato pollen has an outsized role in reproduction, providing essential cellular and genetic material for fertilization and fruit generation. Unfortunately, high temperatures reduce pollination efficiency, harm fruit set and size, and ultimately diminish yield. This project attempts to answer basic questions about pollen growth and function during normal and heat-stressed conditions. Pollen from ~200 genetically diverse tomato and wild relative accessions will be observed as it grows at various temperatures. High-throughput microscopy will be paired with computer vision to phenotype the thousands of image sequences generated by this experiment. By combining pollen growth phenotypes with genome sequence data for all accessions, we plan to identify relevant genomic regions to target for functional description and crop improvement.

The development of new phenomics approaches to image and process data from the subcellular to ecosystem-scale has accelerated over the past decade. Many of these tools are produced “in-house” within a single lab or a group of collaborating labs, making it hard to keep up with the state-of-the-art for phenomics hardware and software development. The Plant Cell Atlas Phenomics Committee is creating a collaborative space that connects phenomics developers with each other and with the greater plant science community, with the goal of facilitating wide-reaching collaborations. To do this, we will be hosting a video series called “How We Built It” where developers provide a short tour of their inventions and relevant biological applications. We will follow the series with a more in-depth networking event where plant scientists can connect with the inventors and discuss collaborative opportunities. Our goal is to streamline the invention of new phenotyping tools and broaden the application of existing tools.

Nitrogen inputs can be an important cost consideration for farmers in terms of economic profit as well as environmental impact. Elucidating genetic regions that are associated with plant phenotype response to nitrogen stress can help in facilitating breeding approaches that can mitigate these costs. A diverse population of 272 maize lines was planted at a field site in Champaign, IL in two consecutive years in reduced nitrogen conditions. 302 phenotypes were recorded including: seed ionomic content, root structural traits derived from 2-dimensional images as well as a 3-dimensional representation generated from x-ray computed tomography (XRT) scans, root traits extrapolated from mini-rhizotron systems, drone images across the growing season and end of season agronomic traits such as biomass and yield. BLUP models were fit to obtain estimates of single year genotypic values as well as across years values both of which took into account year specific spatial variation. Individual years and combined years BLUP values were used as response variables in genome wide association studies (GWAS) to identify loci significantly associated with each set of values. Significant associations were identified for all phenotype categories.

Arsenic (As) pollutes large regions of Asia. Despite phytoremediation initiatives using hyperaccumulators to remove As from contaminated soil, farmers remain reluctant to employ such strategies because of the low biomass and economic value of hyperaccumulating plants. In this study, we demonstrate that cassava can be used for As remediation using a high-throughput root phenotyping platform for cassava roots that we previously developed [1]. Using this phenotyping platform, we identified contrasting root traits associated with As uptake for the two genotypes Rayong 11(R11) and Rayong 90 (R90). Both cassava varieties were grown in pot systems under control (0 mg kg-1 As) and high As (50 mg kg-1 As) conditions and harvested 120 days after planting. We found As stress to reduce shoot and plant dry weight by 57% and 53%, respectively, whereas root dry weight and root traits showed only a slight change. Under As stress, R11 had a 75% higher nodal root number and a 59% lower basal root number than R90. Moreover, R11 root (100 mg kg-1 As) and branch (9 mg kg-1 As) tissues had considerably higher As concentrations than the same tissues in R90. The bioaccumulation coefficient for R11 (2.1) was significantly greater than for R90 (0.9). Additionally, bioethanol yields were unaffected by the presence of As in cassava starch. We suggest that cassava is a promising crop for phytoremediation and that root phenotyping is essential to breed cassava varieties with enhanced As uptake.

We quantify the shape of hooked hairs which is a newly observed phenotype of epidermal cell extensions [1] in the common bean genotype L88-57 (Phaseolus vulgaris). The hooked hairs emerge below-ground before the root hairs and have a distinct ‘hooking’ morphology. We generated a dataset capturing their full distribution under the microscope within 3-5 days of germination. We quantify their shape by a novel computational pipeline that can automatically phenotype morphology. Our phenotyping pipeline quantifies traits like length, curvature, perimeter, area, and ‘hooking.’ Our objective is to quantify their response to nutrient stress to determine the function of hooked hairs in common bean during early development. We used the pipeline for analyzing our dataset of hydroponically grown beans and observed statistically significant responses compared to the control for length, curvature, perimeter, and area to nitrogen (p<0.001**) and phosphorus (p<0.001**) stress treatments. The calculation of ‘hooking’ for our dataset is still ongoing. We are simultaneously developing a landmark-free method for the two-dimensional shape analysis of our dataset and believe that our phenotyping efforts will enable the high-throughput analysis of morphological root hair traits for any plant species.

Quantifying phenotypes of root-root interactions would allow a greater understanding of how plants react to belowground competition through plasticity of architectural traits. Past research has shown that plants will over proliferate roots in the presence of competition, leaving less resources to allocate above ground, negatively impacting shoot growth and yields [1]. Further evidence highlights a response to neighboring plants in the root architecture of Arabidposis thaliana, as individuals concentrate root mass towards their competitors [2]. To visualize and quantify root architecture plasticity involved in these root-root interactions in real soil, we developed a modified mesocosm system. Within the mesocosm box common bean (Phaseolus vulgaris) seeds were germinated 10 inches apart from each other. Mesh screens were placed on either side of each bean, in order to capture root growth towards each other and/or away from each other. Plants were harvested at the 7-week mark, when the root archetype of each individual was developed and prominent. To harvest our boxes three sides of the mesocosm box was removed and the soil was washed away. Images of each box was taken to produce 3D models. Improvements to the mesocosm system will be made to reliably extract root traits in real soil. These experiments will shine light on an understudied section of crop science and will allow farmers and researchers a better understanding of an otherwise unseen phenomena.

The symbiosis between crops and arbuscular mycorrhizal fungi (AMF) have become an attractive route towards achieving carbon neutral agriculture and reducing the use of chemical fertilizers. Yet, our understanding of how active AMF infections influence the uptake, allocation, and exchange of carbon is limited. Here, we combine X-ray CT and PET imaging to observe and quantify the flow of carbon from leaves to roots to hyphae. Comparison of maize grown with and without AMF allows us to measure changes in the amount of 11CO2 taken up in leaves and subsequently the amount of 11C allocated to below-ground roots. Then, co-registered CT and PET images are used to identify hot spots which may indicate active AMF infection sites. Finally, analysis of 11C kinetics at these hot spots are used to assess the amount of carbon exchanged between maize roots and hyphae. By combining structural and biochemical information, we begin to deepen our understanding of the different types of changes in carbon flow in Maize-AMF systems and how we can improve sustainable agriculture efforts.

Plant roots exhibit distinct architectural organization and overall shape. Current concepts to quantify architectural variation assume a homogeneous phenotype for a given genotype. However, this assumption neglects the observable variation in root architecture for two reasons: (i) sampling strategies are designed to capture architectural variation only for the most common phenotype, and (ii) traits are often measured locally within a root system and ignore the architectural organization. Here, we introduce a new concept: the phenotypic spectrum of crop roots to quantify architectural variation as the number of architecture types for one genotype in a specific environment. We use the shape descriptor DS-curve to characterize the whole root system architecture. Using DS curves as a core, we developed a computing pipeline that combines Kmeans++ clustering, outlier filtering and the Fréchet distance as a similarity metric to classify types of root architectures. Subsequently, we applied this pipeline to analyze a field dataset including three common bean (Phaseolus vulgaris) genotypes DOR364 (n=797), L88_57 (n=1772), and SEQ7 (n=768) under non-limiting and water-stressed conditions in 2015 and 2016. We found DOR364 showed five different root architecture types across environments, while L88_57 and SEQ7 showed four. The total variation within classified root architecture types of DOR364, L88_57, and SEQ reduced by 58.59%, 50.19% and 53.01%, compared to the variation of the complete data sets. DOR364 had stable fractions of root architecture types across environments. In contrast, L88_57 and SEQ7 showed more variation in their fractions. There was no significant biomass difference among root architecture types for all studied genotypes within each environment. As such, we hypothesize that the phenotypic spectrum might buffer the impact of environmental stresses as an acclimatization strategy by changing the composition of root architecture types at the population level.

High-throughput phenotyping and genotyping have provided a vast source of information for evaluating the genetic merit of different breeding materials, but their implementation has been limited in alfalfa due to the complexity of the genome and the perennial nature of the crop. Vegetative indices (VIs) collected from an unmanned aerial vehicle (UAV) equipped with multi-spectral camera can be used to study forage growth and development throughout each growth cycle. Random regression models could be implemented to fit such longitudinal phenotypes like VIs collected over time to estimate growth curves, to access genetic variation in growth and the relations of VIs to end-use traits like forage yield and quality. The main objectives of this project are (1) to incorporate aerial high-throughput phenotyping to predict performance and genetic merit of the breeding materials, (2) to fit longitudinal random regression model to estimate genotype-specific growth curves, and (3) to develop a genotyping approach to estimate genetic relationships between alfalfa populations. The imaging of the alfalfa experimental trials was done every ~ 4.3 days throughout the growing season. The Vegetative indices (VIs) close to the harvest date were extracted and used to fit multi-traits models to evaluate the genetic correlations between VIs and forage biomass yield. The VIs considered were Normalized Vegetative index (NDVI), Green NDVI, Red Edge NDVI, simple ratio of Near Infrared to Red (NIR), and Digital Surface Map (DSM). The preliminary results showed highest correlation of Green NDVI and biomass yield (0.4053, 0.7875, and 0.6779), followed by Rededge NDVI and biomass yield (0.417, 0.7898, and 0.6417) for the first, second and third cuttings respectively for the experimental trial located at Helfer, Ithaca. Heritability estimates ranging from 0.03 to 0.75 was observed indicating the presence of genetic variation in these VIs. Pairwise Fst values estimated from population-level genotyping approach was found to be efficient estimates of genetic relatedness between populations. Random regression models with a linear spline function and legendre polynomials including other environmental trials are under evaluation to see the potentiality of these models to fit VIs from multiple time points.

Higher temperatures across the globe are causing an increase in the frequency and severity of droughts. In agricultural crops, this results in reduced yields, financial losses for farmers, and increased food costs at the supermarket. Root architecture plays a major role in a plant’s ability to survive and perform under drought conditions but phenotyping root growth to determine the genetic and environmental factors involved is extremely difficult due to roots being under the soil. RootBot is an automated high-throughput phenotyper that eliminates many of the difficulties and time constraints for performing multiple drought-stress studies. RootBot can simulate plant growth conditions during the first 72 hours of growth using transparent plates filled with soil (as opposed to synthetic media such as agar). RootBot has the capacity for up to 50 plates at a time, however, designing a system to organize these plates, image them at the appropriate times, and save and analyze the data for many plates simultaneously is challenging. To improve upon the pipeline, we incorporate strategies from existing phenotyping pipelines into the imaging and measurement processes. We will also investigate the genotypes using GWAS to identify sequence variants associated with drought tolerance or lack thereof. This pipeline will improve RootBot’s abilities in high-throughput phenotyping and the information gathered will be helpful towards future genetic engineering and breeding.

Canola has a prominent floral signature and requires careful consideration when selecting spectral indices for yield estimation. This study evaluated several spectral indices derived from high-resolution RGB images. A small plot (2.75m x 6m) experiment was conducted at Kernen Research Farm, Saskatoon, where canola was grown under varying row spacings and seeding rates (192 plots). The canopy reflectance was imaged during the flowering period and seed yield was obtained at physiological maturity. Indices were evaluated for accuracy in quantifying canola flowers in high-resolution RGB imagery with within-canopy shadow pixels. Digitalized flower number from the peak flowering date was used to test and validate a non-linear three-parameter asymptotic regression model to simulate canola seed yield. 70 % of the data was used to develop the model, and 30 % was used to validate the model. Model performance was tested with Pseudo-R2, r, MAE, and RMSE. HrFI (High-resolution Flowering Index) and MYI (Modified Yellowness Index) were able to accurately identify flowering pixels with the least amount of error pixel. The yield simulation model resulted in a pseudo-R2 value of 0.11 for the tested model and, a correlation of 0.91 for validation with RMSE and MAE of 343.1 and 265.3, respectively. Our results indicate that the HrFI index is a better indicator of yield potential compared to NDYI as the metric is well capable of handling within canopy shadows. Further studies are necessary to evaluate the performance of HrFI for medium resolution-UAV and satellite imagery.

Plant roots are responsible for essential functions like nutrient uptake, anchorage, and storage. Study of root uptake mechanisms for macro nutrients like nitrogen, phosphorus, potassium, and sulphur is vital to our understanding of their role in plant growth and development. Small signaling peptides (SSPs), are hormones which regulate diverse plant developmental processes including root growth. However, their involvement in regulation of nutrient uptake by roots is poorly understood. We recently developed a hydroponics- based plant growth system which combines ion chromatography with synthetic peptide application, to analyze the depletion rates of nutrients by Medicago truncatula roots. Application of the synthetic SSP MtCEP1 and AtCEP1 led to enhanced uptake of nitrates, sulphates, and phosphates. To further elucidate the molecular mechanism of nutrient uptake mediated by these peptides, we conducted an RNAseq of M. truncatula roots treated with the peptides. A differential gene expression analysis revealed thousands of peptide responsive genes. Several known nitrate transporters and a sulphate transporter AtSULTR3:5-like gene showed enhanced expression under both, MtCEP1 and AtCEP1 peptide application. Multiple, as of yet uncharacterized, CEP peptide responsive pathway regulatory genes such as kinases and transcription factors were also identified through this transcriptomic analysis. This study highlights the potential of phenomics enabled biology to uncover target genes for improving agriculturally important traits such as nutrient uptake.

Anthropogenic factors such as climate change, harsh agricultural practices, and mining have contributed to increases in soil salinization and heavy metal contamination. Highly saline environments drastically lower yield for crop species and elevated levels of toxic metals like cadmium are carcinogenic in the environment. Some plants that have evolved in high-salinity habitats or in soils with heavy metals could be used to remediate contaminated soils. Halophytes are plants with various adaptations that allow them to survive and reproduce in saline conditions. General mechanisms for salt tolerance/uptake in halophytes are hypothesized to help deal with other stresses like heavy metals. Plants with these traits could be utilized to extract salt and heavy metals from affected soils in a process called phytoremediation. We plan to develop Sea Rocket (Cakile maritima) in the Mustard family as a model system to understand mechanisms of salt (NaCl) and cadmium uptake and tolerance, as it has been shown to accumulate both. As part of this study, we will hydroponically grow C. maritima in different stress treatments using salt and cadmium. As the plants uptake the pollutants, the conductivity of the solution will change. We will develop an automated pipeline to track these changes in real-time using conductivity sensors. In addition, we will sample root and leaf tissues at various time points to measure salt and cadmium uptake using ICP-OES elemental analysis. This data will provide insights into salt and cadmium uptake/tolerance and paves a path toward efficient and viable solutions improving phytoremediation approaches.

Understanding root traits is essential to improve water uptake, increase nitrogen capture and raise carbon sequestration from the atmosphere. However, high-throughput phenotyping to quantify root traits for deeper field-grown roots remain a challenge. Recently developed open-source methods use image-based 3D reconstruction algorithms to build 3D models of plant roots from multiple 2D images and can extract root traits and phenotypes. Most of these methods rely on automated image orientation (Structure from Motion)[1] and dense image matching (Multiple View Stereo) algorithms to produce a 3D point cloud or mesh model from 2D images. Until now it is not known how the performance of these methods compares to each other when applied to field-grown roots. We investigate commonly used open-source pipelines on a test panel of twelve contrasting maize genotypes grown in real field conditions in this comparison study [2-6]. We compare 3D point clouds in terms of number of points, computation time, and model surface density. This comparison study will provide insight into the performance of different open-source pipelines for maize root phenotyping, and illuminates trade-offs between 3D model quality and performance cost for future high-throughput 3D root phenotyping.

Generating data has become cheaper and easier, but alone is not sufficient to answer biological questions – data must be analyzed and interpreted. However, many algorithms can create or exacerbate biases (e.g., facial-recognition, ancestry, and disease risk). This necessitates incorporating diverse perspectives to confront both the moral and technical “big data challenges”. To move to a future where this is possible, it is necessary for researchers to develop skills in data management, processing, and analytics. Specifically, the field of plant phenotyping has moved from time consuming hand measurements to the use and development of high-throughput phenotyping. These systems require data-enabled/fluent users, yet academic programs in biology do not provide sufficient data science training. Here we present the Bioinformatics in Plant Science (BIPS) program at the University of Missouri (MU) as a model for training the next generation of data-enabled/fluent scientists. BIPS aims to mentor undergraduate students to build foundational skills in plant biology, research, and computational science. Our program pairs biology and computer science students to address biological questions through computational methods, with many focusing on plant phenotyping methods. The students learn to tackle problems using multidisciplinary approaches, alongside learning how to work in teams while building science communication skills (e.g., professional conferences, research forums, presenting to lawmakers). Through peer learning, BIPS students can understand and incorporate diverse perspectives from both the biological and computational side to address one of NSF’s 10 big ideas: harnessing the data revolution.

Crop pest detection and mitigation remains an extremely challenging task for the farmers. Majority of the pest classification and detection techniques rely on supervised deep learning frameworks that require significant human intervention in labeling the input data, thereby making the down-stream tasks tedious. Therefore, this study presents a self-supervised learning (SSL) approach to classifying 12 types of agricultural insect pests from 9549 RGB images, by leveraging the Bootstrap your own latent (BYOL) algorithm. SSL uses minimal labeling and is indifferent to data augmentations or distortions. Hence, latent representations from pretrained SSL networks could be generalized well for downstream tasks like classification or object detection. For desirable classification of the insect images, the greatest challenges observed were: i) large intra-class variation (the same insect was found with different colors and patterns), and ii) complex background with inconspicuous foreground. Hence, to overcome these issues and aid generalizability of the representations learned through BYOL, entropy-guided segmentation (segments based on texture not color), is proposed as input to the SSL network in this study. Both raw and segmented images were separately fed to two independent BYOL SSL networks, i.e., with ResNet18 and ResNet50 architectures as the backbone. The efficacy of the latent representations for downstream applications was assessed using linear evaluation, and subsequently compared with traditional transfer learning outcomes from ResNet18 and ResNet50. The results indicated that, while ResNet50 backbone intuitively performed better in all cases, SSL aided with entropy-based segmentation led to ~94% classification accuracy compared to raw images (with ~90% maximum accuracy).