Deep Learning in Image-Based Plant Phenotyping.

A major bottleneck in the crop improvement pipeline is our ability to phenotype crops quickly and efficiently. Image-based, high-throughput phenotyping has a number of advantages because it is nondestructive and reduces human labor, but a new challenge arises in extracting meaningful information from large quantities of image data. Deep learning, a type of artificial intelligence, is an approach used to analyze image data and make predictions on unseen images that ultimately reduces the need for human input in computation. Here, we review the basics of deep learning, assessment of deep learning success, examples of applications of deep learning in plant phenomics, best practices, and open challenges. Expected final online publication date for the Annual Review of Plant Biology, Volume 75 is May 2024. Please see http://www.annualreviews.org/page/journal/pubdates for revised estimates.

Natural variation in Brachypodium distachyon responses to combined abiotic stresses.

The demand for agricultural production is becoming more challenging as climate change increases global temperature and the frequency of extreme weather events. This study examines the phenotypic variation of 149 accessions of Brachypodium distachyon under drought, heat, and the combination of stresses. Heat alone causes the largest amounts of tissue damage while the combination of stresses causes the largest decrease in biomass compared to other treatments. Notably, Bd21-0, the reference line for B. distachyon, did not have robust growth under stress conditions, especially the heat and combined drought and heat treatments. The climate of origin was significantly associated with B. distachyon responses to the assessed stress conditions. Additionally, a GWAS found loci associated with changes in plant height and the amount of damaged tissue under stress. Some of these SNPs were closely located to genes known to be involved in responses to abiotic stresses and point to potential causative loci in plant stress response. However, SNPs found to be significantly associated with a response to heat or drought individually are not also significantly associated with the combination of stresses. This, with the phenotypic data, suggests that the effects of these abiotic stresses are not simply additive, and the responses to the combined stresses differ from drought and heat alone.

The plant response to high CO2 levels is heritable and orchestrated by DNA methylation.

Plant responses to abiotic environmental challenges are known to have lasting effects on the plant beyond the initial stress exposure. Some of these lasting effects are transgenerational, affecting the next generation. The plant response to elevated carbon dioxide (CO2 ) levels has been well studied. However, these investigations are typically limited to plants grown for a single generation in a high CO2 environment while transgenerational studies are rare. We aimed to determine transgenerational growth responses in plants after exposure to high CO2 by investigating the direct progeny when returned to baseline CO2 levels. We found that both the flowering plant Arabidopsis thaliana and seedless nonvascular plant Physcomitrium patens continue to display accelerated growth rates in the progeny of plants exposed to high CO2 . We used the model species Arabidopsis to dissect the molecular mechanism and found that DNA methylation pathways are necessary for heritability of this growth response. More specifically, the pathway of RNA-directed DNA methylation is required to initiate methylation and the proteins CMT2 and CMT3 are needed for the transgenerational propagation of this DNA methylation to the progeny plants. Together, these two DNA methylation pathways establish and then maintain a cellular memory to high CO2 exposure.

A protocol for Chenopodium quinoa pollen germination.

BACKGROUND: Quinoa is an increasingly popular seed crop frequently studied for its tolerance to various abiotic stresses as well as its susceptibility to heat. Estimations of quinoa pollen viability through staining methods have resulted in conflicting results. A more effective alternative to stains is to estimate pollen viability through in vitro germination. Here we report a method for in vitro quinoa pollen germination that could be used to understand the impact of various stresses on quinoa fertility and therefore seed yield or to identify male-sterile lines for breeding. RESULTS: A semi-automated method to count germinating pollen was developed in PlantCV, which can be widely used by the community. Pollen collected on day 4 after first anthesis at zeitgeber time 5 was optimum for pollen germination with an average germination of 68% for accession QQ74 (PI 614886). The optimal length of pollen incubation was found to be 48 h, because it maximizes germination rates while minimizing contamination. The pollen germination medium's pH, boric acid, and sucrose concentrations were optimized. The highest germination rates were obtained with 16% sucrose, 0.03% boric acid, 0.007% calcium nitrate, and pH 5.5. This medium was tested on quinoa accessions QQ74, and cherry vanilla with 68%, and 64% germination efficiencies, respectively. CONCLUSIONS: We provide an in vitro pollen germination method for quinoa with average germination rates of 64 and 68% on the two accessions tested. This method is a valuable tool to estimate pollen viability in quinoa, and to test how stress affects quinoa fertility. We also developed an image analysis tool to semi-automate the process of counting germinating pollen. Quinoa produces many new flowers during most of its panicle development period, leading to significant variation in pollen maturity and viability between different flowers of the same panicle. Therefore, collecting pollen at 4 days after first anthesis is very important to collect more uniformly developed pollen and to obtain high germination rates.

Heat stress changes mineral nutrient concentrations in Chenopodium quinoa seed.

Quinoa is a popular seed crop, often consumed for its high nutritional quality. We studied how heat stress in the roots or the shoots of quinoa plants affected the concentrations of 20 elements (aluminum, arsenic, boron, calcium, cadmium, cobalt, copper, iron, potassium, magnesium, manganese, molybdenum, sodium, nickel, phosphorous, rubidium, sulfur, selenium, strontium, and zinc) in quinoa seed. Elemental concentrations in quinoa seed were significantly changed after an 11-day heat treatment during anthesis. The type of panicle (main, secondary, and tertiary) sampled and the type of heat treatment (root only, shoot only, or whole plants) significantly affected elemental profiles in quinoa seed. Plants were also divided into five sections from top to bottom to assess the effect of panicle position on seed elemental profiles. Plant section had an effect on the concentrations of arsenic, iron, and sodium under control conditions and on copper with heat treatment. Overall, the time of panicle development in relation to the time of heat exposure had the largest effect on seed elemental concentrations. Interestingly, the quinoa plants were exposed to heat only during anthesis of the main panicle, but the elemental concentrations of seeds produced after heat treatment ended were still significantly changed, indicating that heat stress has long-lasting effects on quinoa plants. These findings demonstrate how the nutritional quality of quinoa seeds can be changed significantly even by relatively short heat spells.

Heating quinoa shoots results in yield loss by inhibiting fruit production and delaying maturity.

Increasing global temperatures and a growing world population create the need to develop crop varieties that provide higher yields in warmer climates. There is growing interest in expanding quinoa cultivation, because of the ability of quinoa to produce nutritious grain in poor soils, with little water and at high salinity. The main limitation to expanding quinoa cultivation, however, is the susceptibility of quinoa to temperatures above approximately 32°C. This study investigates the phenotypes, genes and mechanisms that may affect quinoa seed yield at high temperatures. Using a differential heating system where only roots or only shoots were heated, quinoa yield losses were attributed to shoot heating. Plants with heated shoots lost 60-85% yield as compared with control plants. Yield losses were the result of lower fruit production, which lowered the number of seeds produced per plant. Furthermore, plants with heated shoots had delayed maturity and greater non-reproductive shoot biomass, whereas plants with both heated roots and heated shoots produced higher yields from the panicles that had escaped the heat, compared with the control. This suggests that quinoa uses a type of avoidance strategy to survive heat. Gene expression analysis identified transcription factors differentially expressed in plants with heated shoots and low yield that had been previously associated with flower development and flower opening. Interestingly, in plants with heated shoots, flowers stayed closed during the day while the control flowers were open. Although a closed flower may protect the floral structures, this could also cause yield losses by limiting pollen dispersal, which is necessary to produce fruit in the mostly female flowers of quinoa.

Classifying cold-stress responses of inbred maize seedlings using RGB imaging.

Increasing the tolerance of maize seedlings to low-temperature episodes could mitigate the effects of increasing climate variability on yield. To aid progress toward this goal, we established a growth chamber-based system for subjecting seedlings of 40 maize inbred genotypes to a defined, temporary cold stress while collecting digital profile images over a 9-daytime course. Image analysis performed with PlantCV software quantified shoot height, shoot area, 14 other morphological traits, and necrosis identified by color analysis. Hierarchical clustering of changes in growth rates of morphological traits and quantification of leaf necrosis over two time intervals resulted in three clusters of genotypes, which are characterized by unique responses to cold stress. For any given genotype, the set of traits with similar growth rates is unique. However, the patterns among traits are different between genotypes. Cold sensitivity was not correlated with the latitude where the inbred varieties were released suggesting potential further improvement for this trait. This work will serve as the basis for future experiments investigating the genetic basis of recovery to cold stress in maize seedlings.

Components of Water Use Efficiency Have Unique Genetic Signatures in the Model C4 Grass Setaria.

Plant growth and water use are interrelated processes influenced by genetically controlled morphological and biochemical characteristics. Improving plant water use efficiency (WUE) to sustain growth in different environments is an important breeding objective that can improve crop yields and enhance agricultural sustainability. However, genetic improvement of WUE using traditional methods has proven difficult due to the low throughput and environmental heterogeneity of field settings. To overcome these limitations, this study utilizes a high-throughput phenotyping platform to quantify plant size and water use of an interspecific Setaria italica × Setaria viridis recombinant inbred line population at daily intervals in both well-watered and water-limited conditions. Our findings indicate that measurements of plant size and water use are correlated strongly in this system; therefore, a linear modeling approach was used to partition this relationship into predicted values of plant size given water use and deviations from this relationship at the genotype level. The resulting traits describing plant size, water use, and WUE all were heritable and responsive to soil water availability, allowing for a genetic dissection of the components of plant WUE under different watering treatments. Linkage mapping identified major loci underlying two different pleiotropic components of WUE. This study indicates that alleles controlling WUE derived from both wild and domesticated accessions can be utilized to predictably modulate trait values given a specified precipitation regime in the model C4 genus Setaria.

Raspberry Pi-powered imaging for plant phenotyping.

PREMISE OF THE STUDY: Image-based phenomics is a powerful approach to capture and quantify plant diversity. However, commercial platforms that make consistent image acquisition easy are often cost-prohibitive. To make high-throughput phenotyping methods more accessible, low-cost microcomputers and cameras can be used to acquire plant image data. METHODS AND RESULTS: We used low-cost Raspberry Pi computers and cameras to manage and capture plant image data. Detailed here are three different applications of Raspberry Pi-controlled imaging platforms for seed and shoot imaging. Images obtained from each platform were suitable for extracting quantifiable plant traits (e.g., shape, area, height, color) en masse using open-source image processing software such as PlantCV. CONCLUSIONS: This protocol describes three low-cost platforms for image acquisition that are useful for quantifying plant diversity. When coupled with open-source image processing tools, these imaging platforms provide viable low-cost solutions for incorporating high-throughput phenomics into a wide range of research programs.

PlantCV v2: Image analysis software for high-throughput plant phenotyping.

Systems for collecting image data in conjunction with computer vision techniques are a powerful tool for increasing the temporal resolution at which plant phenotypes can be measured non-destructively. Computational tools that are flexible and extendable are needed to address the diversity of plant phenotyping problems. We previously described the Plant Computer Vision (PlantCV) software package, which is an image processing toolkit for plant phenotyping analysis. The goal of the PlantCV project is to develop a set of modular, reusable, and repurposable tools for plant image analysis that are open-source and community-developed. Here we present the details and rationale for major developments in the second major release of PlantCV. In addition to overall improvements in the organization of the PlantCV project, new functionality includes a set of new image processing and normalization tools, support for analyzing images that include multiple plants, leaf segmentation, landmark identification tools for morphometrics, and modules for machine learning.

Temporal network analysis identifies early physiological and transcriptomic indicators of mild drought in Brassica rapa.

The dynamics of local climates make development of agricultural strategies challenging. Yield improvement has progressed slowly, especially in drought-prone regions where annual crop production suffers from episodic aridity. Underlying drought responses are circadian and diel control of gene expression that regulate daily variations in metabolic and physiological pathways. To identify transcriptomic changes that occur in the crop Brassica rapa during initial perception of drought, we applied a co-expression network approach to associate rhythmic gene expression changes with physiological responses. Coupled analysis of transcriptome and physiological parameters over a two-day time course in control and drought-stressed plants provided temporal resolution necessary for correlation of network modules with dynamic changes in stomatal conductance, photosynthetic rate, and photosystem II efficiency. This approach enabled the identification of drought-responsive genes based on their differential rhythmic expression profiles in well-watered versus droughted networks and provided new insights into the dynamic physiological changes that occur during drought.

Morphological Plant Modeling: Unleashing Geometric and Topological Potential within the Plant Sciences.

The geometries and topologies of leaves, flowers, roots, shoots, and their arrangements have fascinated plant biologists and mathematicians alike. As such, plant morphology is inherently mathematical in that it describes plant form and architecture with geometrical and topological techniques. Gaining an understanding of how to modify plant morphology, through molecular biology and breeding, aided by a mathematical perspective, is critical to improving agriculture, and the monitoring of ecosystems is vital to modeling a future with fewer natural resources. In this white paper, we begin with an overview in quantifying the form of plants and mathematical models of patterning in plants. We then explore the fundamental challenges that remain unanswered concerning plant morphology, from the barriers preventing the prediction of phenotype from genotype to modeling the movement of leaves in air streams. We end with a discussion concerning the education of plant morphology synthesizing biological and mathematical approaches and ways to facilitate research advances through outreach, cross-disciplinary training, and open science. Unleashing the potential of geometric and topological approaches in the plant sciences promises to transform our understanding of both plants and mathematics.

Cross-species complementation reveals conserved functions for EARLY FLOWERING 3 between monocots and dicots.

Plant responses to the environment are shaped by external stimuli and internal signaling pathways. In both the model plant Arabidopsis thaliana (Arabidopsis) and crop species, circadian clock factors are critical for growth, flowering, and circadian rhythms. Outside of Arabidopsis, however, little is known about the molecular function of clock gene products. Therefore, we sought to compare the function of Brachypodium distachyon (Brachypodium) and Setaria viridis (Setaria) orthologs of EARLY FLOWERING 3, a key clock gene in Arabidopsis. To identify both cycling genes and putative ELF3 functional orthologs in Setaria, a circadian RNA-seq dataset and online query tool (Diel Explorer) were generated to explore expression profiles of Setaria genes under circadian conditions. The function of ELF3 orthologs from Arabidopsis, Brachypodium, and Setaria was tested for complementation of an elf3 mutation in Arabidopsis. We find that both monocot orthologs were capable of rescuing hypocotyl elongation, flowering time, and arrhythmic clock phenotypes. Using affinity purification and mass spectrometry, our data indicate that BdELF3 and SvELF3 could be integrated into similar complexes in vivo as AtELF3. Thus, we find that, despite 180 million years of separation, BdELF3 and SvELF3 can functionally complement loss of ELF3 at the molecular and physiological level.

Natural variation in the C-repeat binding factor cold response pathway correlates with local adaptation of Arabidopsis ecotypes.

The natural range of Arabidopsis thaliana (Arabidopsis) encompasses geographical regions that have greatly differing local climates, including harshness of winter temperatures. A question thus raised is whether differences in freezing tolerance might contribute to local adaptation in Arabidopsis. Consistent with this possibility is that Arabidopsis accessions differ in freezing tolerance and that those collected from colder northern latitudes are generally more tolerant to freezing than those collected from warmer southern latitudes. Moreover, recent studies with Arabidopsis genotypes collected from sites in Sweden (SW) and Italy (IT) have established that the two accessions are locally adapted, that the SW ecotype is more tolerant of freezing than the IT ecotype, and that genetic differences between the two ecotypes that condition local adaptation and freezing tolerance map to a region that includes the C-repeat binding factor (CBF) locus. The CBF locus includes three genes - CBF1, CBF2 and CBF3 - that are induced by low temperature and encode transcription factors that regulate a group of more than 100 genes, the CBF regulon, which impart freezing tolerance. Here we show that cold induction of most CBF regulon genes is lower in IT plants compared with SW plants, and that this is due to the IT CBF2 gene encoding a non-functional CBF2 protein. The non-functional IT CBF2 protein also contributes to the lower freezing tolerance of the IT plants compared with the SW plants. Taken together, studies on the SW and IT ecotypes provide evidence that natural variation in the CBF pathway has contributed to adaptive evolution in these Arabidopsis populations.

A Versatile Phenotyping System and Analytics Platform Reveals Diverse Temporal Responses to Water Availability in Setaria.

Phenotyping has become the rate-limiting step in using large-scale genomic data to understand and improve agricultural crops. Here, the Bellwether Phenotyping Platform for controlled-environment plant growth and automated multimodal phenotyping is described. The system has capacity for 1140 plants, which pass daily through stations to record fluorescence, near-infrared, and visible images. Plant Computer Vision (PlantCV) was developed as open-source, hardware platform-independent software for quantitative image analysis. In a 4-week experiment, wild Setaria viridis and domesticated Setaria italica had fundamentally different temporal responses to water availability. While both lines produced similar levels of biomass under limited water conditions, Setaria viridis maintained the same water-use efficiency under water replete conditions, while Setaria italica shifted to less efficient growth. Overall, the Bellwether Phenotyping Platform and PlantCV software detected significant effects of genotype and environment on height, biomass, water-use efficiency, color, plant architecture, and tissue water status traits. All ∼ 79,000 images acquired during the course of the experiment are publicly available.

Lights, camera, action: high-throughput plant phenotyping is ready for a close-up.

Anticipated population growth, shifting demographics, and environmental variability over the next century are expected to threaten global food security. In the face of these challenges, crop yield for food and fuel must be maintained and improved using fewer input resources. In recent years, genetic tools for profiling crop germplasm has benefited from rapid advances in DNA sequencing, and now similar advances are needed to improve the throughput of plant phenotyping. We highlight recent developments in high-throughput plant phenotyping using robotic-assisted imaging platforms and computer vision-assisted analysis tools.

Transcriptional networks-crops, clocks, and abiotic stress.

Several factors affect the yield potential and geographical range of crops including the circadian clock, water availability, and seasonal temperature changes. In order to sustain and increase plant productivity on marginal land in the face of both biotic and abiotic stresses, we need to more efficiently generate stress-resistant crops through marker-assisted breeding, genetic modification, and new genome-editing technologies. To leverage these strategies for producing the next generation of crops, future transcriptomic data acquisition should be pursued with an appropriate temporal design and analyzed with a network-centric approach. The following review focuses on recent developments in abiotic stress transcriptional networks in economically important crops and will highlight the utility of correlation-based network analysis and applications.