Advanced deep learning models for phenotypic trait extraction and cultivar classification in lychee using photon-counting micro-CT imaging

Introduction

In contemporary agronomic research, the focus has increasingly shifted towards non-destructive imaging and precise phenotypic characterization. A photon-counting micro-CT system has been developed, which is capable of imaging lychee fruit at the micrometer level and capturing a full energy spectrum, thanks to its advanced photon-counting detectors.

Methods

For automatic measurement of phenotypic traits, seven CNN-based deep learning models including AttentionUNet, DeeplabV3+, SegNet, TransUNet, UNet, UNet++, and UNet3+ were developed. Machine learning techniques tailored for small-sample training were employed to identify key characteristics of various lychee species.

Results

These models demonstrate outstanding performance with Dice, Recall, and Precision indices predominantly ranging between 0.90 and 0.99. The Mean Intersection over Union (MIoU) consistently falls between 0.88 and 0.98. This approach served both as a feature selection process and a means of classification, significantly enhancing the study's ability to discern and categorize distinct lychee varieties.

Discussion

This research not only contributes to the advancement of non-destructive plant analysis but also opens new avenues for exploring the intricate phenotypic variations within plant species.

Meta-omics integration approach reveals the effect of soil native microbiome diversity in the performance of inoculant Azospirillum brasilense

Plant growth promoting bacteria (PGPB) have been used as integrative inputs to minimize the use of chemical fertilizers. However, a holistic comprehension about PGPB-plant-microbiome interactions is still incipient. Furthermore, the interaction among PGPB and the holobiont (host-microbiome association) represent a new frontier to plant breeding programs. We aimed to characterize maize bulk soil and rhizosphere microbiomes in irradiated soil (IS) and a native soil (NS) microbial community gradient (dilution-to-extinction) with Azospirillum brasilense Ab-V5, a PGPB commercial inoculant. Our hypothesis was that plant growth promotion efficiency is a result of PGPB niche occupation and persistence according to the holobiont conditions. The effects of Ab-V5 and NS microbial communities were evaluated in microcosms by a combined approach of microbiomics (species-specific qPCR, 16S rRNA metataxonomics and metagenomics) and plant phenomics (conventional and high-throughput methods). Our results revealed a weak maize growth promoting effect of Ab-V5 inoculation in undiluted NS, contrasting the positive effects of NS dilutions 10^-3, 10^-6, 10^-9 and IS with Ab-V5. Alpha diversity in NS + Ab-V5 soil samples was higher than in all other treatments in a time course of 25 days after sowing (DAS). At 15 DAS, alpha diversity indexes were different between NS and IS, but similar in all NS dilutions in rhizospheric samples. These differences were not persistent at 25 DAS, demonstrating a stabilization process in the rhizobiomes. In NS 10^-3 +Ab-V5 and NS 10^-6 Ab-V5, Ab-V5 persisted in the maize rhizosphere until 15 DAS in higher abundances compared to NS. In NS + Ab-V5, abundance of six taxa were positively correlated with response to (a)biotic stresses in plant-soil interface. Genes involved in bacterial metabolism of riboses and amino acids, and cresol degradation were abundant on NS 10^-3 + Ab-V5, indicating that these pathways can contribute to plant growth promotion and might be a result of Ab-V5 performance as a microbial recruiter of beneficial functions to the plant. Our results demonstrated the effects of holobiont on Ab-V5 performance. The meta-omics integration supported by plant phenomics opens new perspectives to better understanding of inoculants-holobiont interaction and for developing better strategies for optimization in the use of microbial products.

Hyperspectral reflectance imaging for nondestructive evaluation of root rot in Korean ginseng (Panax ginseng Meyer)

Root rot of Panax ginseng caused by Cylindrocarpon destructans, a soil-borne fungus is typically diagnosed by frequently checking the ginseng plants or by evaluating soil pathogens in a farm, which is a time- and cost-intensive process. Because this disease causes huge economic losses to ginseng farmers, it is important to develop reliable and non-destructive techniques for early disease detection. In this study, we developed a non-destructive method for the early detection of root rot. For this, we used crop phenotyping and analyzed biochemical information collected using the HSI technique. Soil infected with root rot was divided into sterilized and infected groups and seeded with 1-year-old ginseng plants. HSI data were collected four times during weeks 7-10 after sowing. The spectral data were analyzed and the main wavelengths were extracted using partial least squares discriminant analysis. The average model accuracy was 84% in the visible/near-infrared region (29 main wavelengths) and 95% in the short-wave infrared (19 main wavelengths). These results indicated that root rot caused a decrease in nutrient absorption, leading to a decline in photosynthetic activity and the levels of carotenoids, starch, and sucrose. Wavelengths related to phenolic compounds can also be utilized for the early prediction of root rot. The technique presented in this study can be used for the early and timely detection of root rot in ginseng in a non-destructive manner.

Nondestructive 3D phenotyping method of passion fruit based on X-ray micro-computed tomography and deep learning

Passion fruit is a tropical liana of the Passiflora family that is commonly planted throughout the world due to its abundance of nutrients and industrial value. Researchers are committed to exploring the relationship between phenotype and genotype to promote the improvement of passion fruit varieties. However, the traditional manual phenotyping methods have shortcomings in accuracy, objectivity, and measurement efficiency when obtaining large quantities of personal data on passion fruit, especially internal organization data. This study selected samples of passion fruit from three widely grown cultivars, which differed significantly in fruit shape, size, and other morphological traits. A Micro-CT system was developed to perform fully automated nondestructive imaging of the samples to obtain 3D models of passion fruit. A designed label generation method and segmentation method based on U-Net model were used to distinguish different tissues in the samples. Finally, fourteen traits, including fruit volume, surface area, length and width, sarcocarp volume, pericarp thickness, and traits of fruit type, were automatically calculated. The experimental results show that the segmentation accuracy of the deep learning model reaches more than 0.95. Compared with the manual measurements, the mean absolute percentage error of the fruit width and length measurements by the Micro-CT system was 1.94% and 2.89%, respectively, and the squares of the correlation coefficients were 0.96 and 0.93. It shows that the measurement accuracy of external traits of passion fruit is comparable to manual operations, and the measurement of internal traits is more reliable because of the nondestructive characteristics of our method. According to the statistical data of the whole samples, the Pearson analysis method was used, and the results indicated specific correlations among fourteen phenotypic traits of passion fruit. At the same time, the results of the principal component analysis illustrated that the comprehensive quality of passion fruit could be scored using this method, which will help to screen for high-quality passion fruit samples with large sizes and high sarcocarp content. The results of this study will firstly provide a nondestructive method for more accurate and efficient automatic acquisition of comprehensive phenotypic traits of passion fruit and have the potential to be extended to more fruit crops. The preliminary study of the correlation between the characteristics of passion fruit can also provide a particular reference value for molecular breeding and comprehensive quality evaluation.

A non-destructive coconut fruit and seed traits extraction method based on Micro-CT and deeplabV3+ model

With the completion of the coconut gene map and the gradual improvement of related molecular biology tools, molecular marker-assisted breeding of coconut has become the next focus of coconut breeding, and accurate coconut phenotypic traits measurement will provide technical support for screening and identifying the correspondence between genotype and phenotype. A Micro-CT system was developed to measure coconut fruits and seeds automatically and nondestructively to acquire the 3D model and phenotyping traits. A deeplabv3+ model with an Xception backbone was used to segment the sectional image of coconut fruits and seeds automatically. Compared with the structural-light system measurement, the mean absolute percentage error of the fruit volume and surface area measurements by the Micro-CT system was 1.87% and 2.24%, respectively, and the squares of the correlation coefficients were 0.977 and 0.964, respectively. In addition, compared with the manual measurements, the mean absolute percentage error of the automatic copra weight and total biomass measurements was 8.85% and 25.19%, respectively, and the adjusted squares of the correlation coefficients were 0.922 and 0.721, respectively. The Micro-CT system can nondestructively obtain up to 21 agronomic traits and 57 digital traits precisely.

Proximal and remote sensing in plant phenomics: 20 years of progress, challenges, and perspectives

Plant phenomics (PP) has been recognized as a bottleneck in studying the interactions of genomics and environment on plants, limiting the progress of smart breeding and precise cultivation. High-throughput plant phenotyping is challenging owing to the spatio-temporal dynamics of traits. Proximal and remote sensing (PRS) techniques are increasingly used for plant phenotyping because of their advantages in multi-dimensional data acquisition and analysis. Substantial progress of PRS applications in PP has been observed over the last two decades and is analyzed here from an interdisciplinary perspective based on 2972 publications. This progress covers most aspects of PRS application in PP, including patterns of global spatial distribution and temporal dynamics, specific PRS technologies, phenotypic research fields, working environments, species, and traits. Subsequently, we demonstrate how to link PRS to multi-omics studies, including how to achieve multi-dimensional PRS data acquisition and processing, how to systematically integrate all kinds of phenotypic information and derive phenotypic knowledge with biological significance, and how to link PP to multi-omics association analysis. Finally, we identify three future perspectives for PRS-based PP: (1) strengthening the spatial and temporal consistency of PRS data, (2) exploring novel phenotypic traits, and (3) facilitating multi-omics communication.

From data to knowledge - big data needs stewardship, a plant phenomics perspective

The research data life cycle from project planning to data publishing is an integral part of current research. Until the last decade, researchers were responsible for all associated phases in addition to the actual research and were assisted only at certain points by IT or bioinformaticians. Starting with advances in sequencing, the automation of analytical methods in all life science fields, including in plant phenotyping, has led to ever-increasing amounts of ever more complex data. The tasks associated with these challenges now often exceed the expertise of and infrastructure available to scientists, leading to an increased risk of data loss over time. The IPK Gatersleben has one of the world's largest germplasm collections and two decades of experience in crop plant research data management. In this article we show how challenges in modern, data-driven research can be addressed by data stewards. Based on concrete use cases, data management processes and best practices from plant phenotyping, we describe which expertise and skills are required and how data stewards as an integral actor can enhance the quality of a necessary digital transformation in progressive research.

High-Throughput Phenotyping Approach for the Evaluation of Heat Stress in Korean Ginseng (Panax ginseng Meyer) Using a Hyperspectral Reflectance Image

Panax ginseng has been used as a traditional medicine to strengthen human health for centuries. Over the last decade, significant agronomical progress has been made in the development of elite ginseng cultivars, increasing their production and quality. However, as one of the significant environmental factors, heat stress remains a challenge and poses a significant threat to ginseng plants’ growth and sustainable production. This study was conducted to investigate the phenotype of ginseng leaves under heat stress using hyperspectral imaging (HSI). A visible/near-infrared (Vis/NIR) and short-wave infrared (SWIR) HSI system were used to acquire hyperspectral images for normal and heat stress-exposed plants, showing their susceptibility (Chunpoong) and resistibility (Sunmyoung and Sunil). The acquired hyperspectral images were analyzed using the partial least squares-discriminant analysis (PLS-DA) technique, combining the variable importance in projection and successive projection algorithm methods. The correlation of each group was verified using linear discriminant analysis. The developed models showed 12 bands over 79.2% accuracy in Vis/NIR and 18 bands with over 98.9% accuracy at SWIR in validation data. The constructed beta-coefficient allowed the observation of the key wavebands and peaks linked to the chlorophyll, nitrogen, fatty acid, sugar and protein content regions, which differentiated normal and stressed plants. This result shows that the HSI with the PLS-DA technique significantly differentiated between the heat-stressed susceptibility and resistibility of ginseng plants with high accuracy.

Review: Application of Artificial Intelligence in Phenomics

Plant phenomics has been rapidly advancing over the past few years. This advancement is attributed to the increased innovation and availability of new technologies which can enable the high-throughput phenotyping of complex plant traits. The application of artificial intelligence in various domains of science has also grown exponentially in recent years. Notably, the computer vision, machine learning, and deep learning aspects of artificial intelligence have been successfully integrated into non-invasive imaging techniques. This integration is gradually improving the efficiency of data collection and analysis through the application of machine and deep learning for robust image analysis. In addition, artificial intelligence has fostered the development of software and tools applied in field phenotyping for data collection and management. These include open-source devices and tools which are enabling community driven research and data-sharing, thereby availing the large amounts of data required for the accurate study of phenotypes. This paper reviews more than one hundred current state-of-the-art papers concerning AI-applied plant phenotyping published between 2010 and 2020. It provides an overview of current phenotyping technologies and the ongoing integration of artificial intelligence into plant phenotyping. Lastly, the limitations of the current approaches/methods and future directions are discussed.

Use of Phenomics for Differentiation of Mungbean (Vigna radiata L. Wilczek) Genotypes Varying in Growth Rates Per Unit of Water

In the human diet, particularly for most of the vegetarian population, mungbean (Vigna radiata L. Wilczek) is an inexpensive and environmentally friendly source of protein. Being a short-duration crop, mungbean fits well into different cropping systems dominated by staple food crops such as rice and wheat. Hence, knowing the growth and production pattern of this important legume under various soil moisture conditions gains paramount significance. Toward that end, 24 elite mungbean genotypes were grown with and without water stress for 25 days in a controlled environment. Top view and side view (two) images of all genotypes captured by a high-resolution camera installed in the high-throughput phenomics were analyzed to extract the pertinent parameters associated with plant features. We tested eight different multivariate models employing machine learning algorithms to predict fresh biomass from different features extracted from the images of diverse genotypes in the presence and absence of soil moisture stress. Based on the mean absolute error (MAE), root mean square error (RMSE), and R squared (R ²) values, which are used to assess the precision of a model, the partial least square (PLS) method among the eight models was selected for the prediction of biomass. The predicted biomass was used to compute the plant growth rates and water-use indices, which were found to be highly promising surrogate traits as they could differentiate the response of genotypes to soil moisture stress more effectively. To the best of our knowledge, this is perhaps the first report stating the use of a phenomics method as a promising tool for assessing growth rates and also the productive use of water in mungbean crop.

Imaging of multiple fluorescent proteins in canopies enables synthetic biology in plants

Reverse genetics approaches have revolutionized plant biology and agriculture. Phenomics has the prospect of bridging plant phenotypes with genes, including transgenes, to transform agricultural fields. Genetically encoded fluorescent proteins (FPs) have revolutionized plant biology paradigms in gene expression, protein trafficking and plant physiology. While the first instance of plant canopy imaging of green fluorescent protein (GFP) was performed over 25 years ago, modern phenomics has largely ignored fluorescence as a transgene expression device despite the burgeoning FP colour palette available to plant biologists. Here, we show a new platform for stand-off imaging of plant canopies expressing a wide variety of FP genes. The platform-the fluorescence-inducing laser projector (FILP)-uses an ultra-low-noise camera to image a scene illuminated by compact diode lasers of various colours, coupled with emission filters to resolve individual FPs, to phenotype transgenic plants expressing FP genes. Each of the 20 FPs screened in plants were imaged at >3 m using FILP in a laboratory-based laser range. We also show that pairs of co-expressed fluorescence proteins can be imaged in canopies. The FILP system enabled a rapid synthetic promoter screen: starting from 2000 synthetic promoters transfected into protoplasts to FILP-imaged agroinfiltrated Nicotiana benthamiana plants in a matter of weeks, which was useful to characterize a water stress-inducible synthetic promoter. FILP canopy imaging was also accomplished for stably transformed GFP potato and in a split-GFP assay, which illustrates the flexibility of the instrument for analysing fluorescence signals in plant canopies.

Combining high-throughput micro-CT-RGB phenotyping and genome-wide association study to dissect the genetic architecture of tiller growth in rice

Manual phenotyping of rice tillers is time consuming and labor intensive, and lags behind the rapid development of rice functional genomics. Thus, automated, non-destructive methods of phenotyping rice tiller traits at a high spatial resolution and high throughput for large-scale assessment of rice accessions are urgently needed. In this study, we developed a high-throughput micro-CT-RGB imaging system to non-destructively extract 739 traits from 234 rice accessions at nine time points. We could explain 30% of the grain yield variance from two tiller traits assessed in the early growth stages. A total of 402 significantly associated loci were identified by genome-wide association study, and dynamic and static genetic components were found across the nine time points. A major locus associated with tiller angle was detected at time point 9, which contained a major gene, TAC1. Significant variants associated with tiller angle were enriched in the 3'-untranslated region of TAC1. Three haplotypes for the gene were found, and rice accessions containing haplotype H3 displayed much smaller tiller angles. Further, we found two loci containing associations with both vigor-related traits identified by high-throughput micro-CT-RGB imaging and yield. The superior alleles would be beneficial for breeding for high yield and dense planting.

Morphological Characterization of Root System Architecture in Diverse Tomato Genotypes during Early Growth

Plant roots exploit morphological plasticity to adapt and respond to different soil environments. We characterized the root system architecture of nine wild tomato species and four cultivated tomato (Solanum lycopersicum L.) varieties during early growth in a controlled environment. Additionally, the root system architecture of six near-isogenic lines from the tomato 'Micro-Tom' mutant collection was also studied. These lines were affected in key genes of ethylene, abscisic acid, and anthocyanin pathways. We found extensive differences between the studied lines for a number of meaningful morphological traits, such as lateral root distribution, lateral root length or adventitious root development, which might represent adaptations to local soil conditions during speciation and subsequent domestication. Taken together, our results provide a general quantitative framework for comparing root system architecture in tomato seedlings and other related species.

High Throughput Determination of Plant Height, Ground Cover, and Above-Ground Biomass in Wheat with LiDAR

Crop improvement efforts are targeting increased above-ground biomass and radiation-use efficiency as drivers for greater yield. Early ground cover and canopy height contribute to biomass production, but manual measurements of these traits, and in particular above-ground biomass, are slow and labor-intensive, more so when made at multiple developmental stages. These constraints limit the ability to capture these data in a temporal fashion, hampering insights that could be gained from multi-dimensional data. Here we demonstrate the capacity of Light Detection and Ranging (LiDAR), mounted on a lightweight, mobile, ground-based platform, for rapid multi-temporal and non-destructive estimation of canopy height, ground cover and above-ground biomass. Field validation of LiDAR measurements is presented. For canopy height, strong relationships with LiDAR (r2 of 0.99 and root mean square error of 0.017 m) were obtained. Ground cover was estimated from LiDAR using two methodologies: red reflectance image and canopy height. In contrast to NDVI, LiDAR was not affected by saturation at high ground cover, and the comparison of both LiDAR methodologies showed strong association (r2 = 0.92 and slope = 1.02) at ground cover above 0.8. For above-ground biomass, a dedicated field experiment was performed with destructive biomass sampled eight times across different developmental stages. Two methodologies are presented for the estimation of biomass from LiDAR: 3D voxel index (3DVI) and 3D profile index (3DPI). The parameters involved in the calculation of 3DVI and 3DPI were optimized for each sample event from tillering to maturity, as well as generalized for any developmental stage. Individual sample point predictions were strong while predictions across all eight sample events, provided the strongest association with biomass (r2 = 0.93 and r2 = 0.92) for 3DPI and 3DVI, respectively. Given these results, we believe that application of this system will provide new opportunities to deliver improved genotypes and agronomic interventions via more efficient and reliable phenotyping of these important traits in large experiments.

Cell to whole-plant phenotyping: the best is yet to come

Imaging and image processing have revolutionized plant phenotyping and are now a major tool for phenotypic trait measurement. Here we review plant phenotyping systems by examining three important characteristics: throughput, dimensionality, and resolution. First, whole-plant phenotyping systems are highlighted together with advances in automation that enable significant throughput increases. Organ and cellular level phenotyping and its tools, often operating at a lower throughput, are then discussed as a means to obtain high-dimensional phenotypic data at elevated spatial and temporal resolution. The significance of recent developments in sensor technologies that give access to plant morphology and physiology-related traits is shown. Overall, attention is focused on spatial and temporal resolution because these are crucial aspects of imaging procedures in plant phenotyping systems.