PRGdb 4.0: an updated database dedicated to genes involved in plant disease resistance process.

The Plant Resistance Genes database (PRGdb; http://prgdb.org/prgdb4/) has been greatly expanded, keeping pace with the increasing amount of available knowledge and data (sequenced proteomes, cloned genes, public analysis data, etc.). The easy-to-use style of the database website has been maintained, while an updated prediction tool, more data and a new section have been added. This new section will contain plant resistance transcriptomic experiments, providing additional easy-to-access experimental information. DRAGO3, the tool for automatic annotation and prediction of plant resistance genes behind PRGdb, has been improved in both accuracy and sensitivity, leading to more reliable predictions. PRGdb offers 199 reference resistance genes and 586.652 putative resistance genes from 182 sequenced proteomes. Compared to the previous release, PRGdb 4.0 has increased the number of reference resistance genes from 153 to 199, the number of putative resistance genes from 177K from 76 proteomes to 586K from 182 sequenced proteomes. A new section has been created that collects plant-pathogen transcriptomic data for five species of agricultural interest. Thereby, with these improvements and data expansions, PRGdb 4.0 aims to serve as a reference to the plant scientific community and breeders worldwide, helping to further study plant resistance mechanisms that contribute to fighting pathogens.

A random forest model for the classification of wheat and rye leaf rust symptoms based on pure spectra at leaf scale.

The pure spectra acquisition of plant disease symptoms is essential to improving the reliability of remote sensing methods in crop protection. The reflectance values read from the pure spectra can be used as valuable training data for development of algorithms designed for plant disease detection at leaf and canopy scale. The aim of this paper is to identify and distinguish spectrally the leaf rust symptoms caused by two closely related special forms (f. sp.) of Puccinia recondita f. sp. tritici on wheat and Puccinia recondita f. sp. recondita on rye at leaf scale. Spectral measurements were made with FieldSpec 3 spectrometer in the wavelength range of 350-2500 nm. The spectrometer was connected to a microscope by optical fiber. Raw spectra of uredinia, chlorotic discoloration, green leaves, senescent inoculated leaves and senescent uninoculated leaves of wheat and rye, all of which obtained for this study, were investigated with a view towards making an automized classification of plant species and their phases. The created Random Forest models were tested separately using pure spectra, and from these vegetation indices were derived as predictors. Three vegetation indices, namely CRI, PRI and GNDVI, appeared to be the most robust in terms of distinguishing uredinia from other symptoms on rye and wheat leaves. PRI, EVI, NDVI705, and GNDVI were the most suitable for distinguishing uredinia, chlorotic discoloration, and green leaf stages on rye. That tusk on wheat leaves can be recognized if seven indices (PRI, MSAWI, SAVI, NDVI, NDVI705, GNDVI and RVI) are used together. For the classification of all disease symptoms for both plant species, the most useful were wavelengths in the VIS range: 431-436, 696-703 and 646-686 nm. However, the ranges of SWIR wavelengths (1938, 1955) and NIR wavelengths (1099-1104) also have a high contribution to the discrimination accuracy of the model. In the classification of all disease symptoms, the most important vegetation indices were CRI, OSAVI, and GNDVI. Analysis of the results revealed the advantage of the model based on the selected spectral wavelengths (Hit Rate of 96.6%) in comparison with predictions based on vegetation indices alone (Hit Rate of 91.7%). Both approaches show the highly applicable character of utilizing high quality spectral products such as satellite images in reducing operational costs of crop protection.

Plant Disease Recognition: A Large-Scale Benchmark Dataset and a Visual Region and Loss Reweighting Approach.

Plant disease diagnosis is very critical for agriculture due to its importance for increasing crop production. Recent advances in image processing offer us a new way to solve this issue via visual plant disease analysis. However, there are few works in this area, not to mention systematic researches. In this paper, we systematically investigate the problem of visual plant disease recognition for plant disease diagnosis. Compared with other types of images, plant disease images generally exhibit randomly distributed lesions, diverse symptoms and complex backgrounds, and thus are hard to capture discriminative information. To facilitate the plant disease recognition research, we construct a new large-scale plant disease dataset with 271 plant disease categories and 220,592 images. Based on this dataset, we tackle plant disease recognition via reweighting both visual regions and loss to emphasize diseased parts. We first compute the weights of all the divided patches from each image based on the cluster distribution of these patches to indicate the discriminative level of each patch. Then we allocate the weight to each loss for each patch-label pair during weakly-supervised training to enable discriminative disease part learning. We finally extract patch features from the network trained with loss reweighting, and utilize the LSTM network to encode the weighed patch feature sequence into a comprehensive feature representation. Extensive evaluations on this dataset and another public dataset demonstrate the advantage of the proposed method. We expect this research will further the agenda of plant disease recognition in the community of image processing.

Changes in global Orchidaceae disease geographical research trends: recent incidences, distributions, treatment, and challenges.

Many of the Orchidaceae species are threatened due to environmental changes and over exploitation for full fill global demands. The main objective of this article was critically analyzed the recent global distribution of Orchidaceae diversity, its disease patterns, microbial disease identification, detection, along with prevention and challenges. Critical analysis findings revealed that Orchidaceae growth and developments were affected indirectly or directly as a result of complex microbial ecological interactions. Studies have identified many species associated with orchids, some are pathogenic and cause symptoms such as soft rot, brown rot, brown spot, black rot, wilt, foliar, root rot, anthracnose, leaf spot. The review was provided the comprehensive data to evaluate the identification and detection of microbial disease, which is the most important challenge for sustainable cultivation of Orchidaceae diversity. Furthermore, this article is the foremost of disease triggering microbes, orchid relations, and assimilates various consequences that both promoted the considerate and facts of such disease multipart, and will permit the development of best operative disease management practices.

Real-time plant health assessment via implementing cloud-based scalable transfer learning on AWS DeepLens.

The control of plant leaf diseases is crucial as it affects the quality and production of plant species with an effect on the economy of any country. Automated identification and classification of plant leaf diseases is, therefore, essential for the reduction of economic losses and the conservation of specific species. Various Machine Learning (ML) models have previously been proposed to detect and identify plant leaf disease; however, they lack usability due to hardware sophistication, limited scalability and realistic use inefficiency. By implementing automatic detection and classification of leaf diseases in fruit trees (apple, grape, peach and strawberry) and vegetable plants (potato and tomato) through scalable transfer learning on Amazon Web Services (AWS) SageMaker and importing it into AWS DeepLens for real-time functional usability, our proposed DeepLens Classification and Detection Model (DCDM) addresses such limitations. Scalability and ubiquitous access to our approach is provided by cloud integration. Our experiments on an extensive image data set of healthy and unhealthy fruit trees and vegetable plant leaves showed 98.78% accuracy with a real-time diagnosis of diseases of plant leaves. To train DCDM deep learning model, we used forty thousand images and then evaluated it on ten thousand images. It takes an average of 0.349s to test an image for disease diagnosis and classification using AWS DeepLens, providing the consumer with disease information in less than a second.

Large-spored Drechslera gigantea is a Bipolaris species causing disease on the invasive grass Microstegium vimineum.

Environmentally damaging invasive plants can also serve as reservoir hosts for agricultural pathogens. Microstegium vimineum is an invasive C4 annual grass that is present throughout the midwestern and eastern United States. It can reach high densities in disturbed areas such as crop-forest interfaces, which creates the potential for pathogen spillover from M. vimineum to agricultural crops and native plants. A previous study that surveyed disease on M. vimineum found a large-spored Bipolaris species that was widespread on M. vimineum and also isolated from co-occurring native grasses. Here, we report that the large-spored fungus isolated from M. vimineum and the native grass Elymus virginicus is Drechslera gigantea, based on comparison with published descriptions of morphological traits, and establish that D. gigantea is a pathogen of M. vimineum and E. virginicus. We review the phylogenetic placement and taxonomic history of D. gigantea and propose that it be reassigned to the genus Bipolaris as Bipolaris gigantea.

Disease Classification in Eggplant Using Pre-trained VGG16 and MSVM.

Currently, the application of deep learning in crop disease classification is one of the active areas of research for which an image dataset is required. Eggplant (Solanum melongena) is one of the important crops, but it is susceptible to serious diseases which hinder its production. Surprisingly, so far no dataset is available for the diseases in this crop. The unavailability of the dataset for these diseases motivated the authors to create a standard dataset in laboratory and field conditions for five major diseases. Pre-trained Visual Geometry Group 16 (VGG16) architecture has been used and the images have been converted to other color spaces namely Hue Saturation Value (HSV), YCbCr and grayscale for evaluation. Results show that the dataset created with RGB and YCbCr images in field condition was promising with a classification accuracy of 99.4%. The dataset also has been evaluated with other popular architectures and compared. In addition, VGG16 has been used as feature extractor from 8th convolution layer and these features have been used for classifying diseases employing Multi-Class Support Vector Machine (MSVM). The analysis depicted an equivalent or in some cases produced better accuracy. Possible reasons for variation in interclass accuracy and future direction have been discussed.

Symptom evolution following the emergence of maize streak virus.

For pathogens infecting single host species evolutionary trade-offs have previously been demonstrated between pathogen-induced mortality rates and transmission rates. It remains unclear, however, how such trade-offs impact sub-lethal pathogen-inflicted damage, and whether these trade-offs even occur in broad host-range pathogens. Here, we examine changes over the past 110 years in symptoms induced in maize by the broad host-range pathogen, maize streak virus (MSV). Specifically, we use the quantified symptom intensities of cloned MSV isolates in differentially resistant maize genotypes to phylogenetically infer ancestral symptom intensities and check for phylogenetic signal associated with these symptom intensities. We show that whereas symptoms reflecting harm to the host have remained constant or decreased, there has been an increase in how extensively MSV colonizes the cells upon which transmission vectors feed. This demonstrates an evolutionary trade-off between amounts of pathogen-inflicted harm and how effectively viruses position themselves within plants to enable onward transmission.

A review of plant leaf fungal diseases and its environment speciation.

There is increasing difficulty in identifying new plant leaf diseases as a result of environmental change. There is a need to identify the factors influencing the emergence and the increasing incidences of these diseases. Here, we present emerging fungal plant leaf diseases and describe their environmental speciation. We considered the factors controlling for local adaptation associated with environmental speciation. We determined that the advent of emergent fungal leaf diseases is closely connected to environmental speciation. Fungal pathogens targeting the leaves may adversely affect the entire plant body. To mitigate the injury caused by these pathogens, it is necessary to be able to detect and identify them early in the infection process. In this way, their distribution, virulence, incidence, and severity could be attenuated.

MARPLE, a point-of-care, strain-level disease diagnostics and surveillance tool for complex fungal pathogens.

BACKGROUND: Effective disease management depends on timely and accurate diagnosis to guide control measures. The capacity to distinguish between individuals in a pathogen population with specific properties such as fungicide resistance, toxin production and virulence profiles is often essential to inform disease management approaches. The genomics revolution has led to technologies that can rapidly produce high-resolution genotypic information to define individual variants of a pathogen species. However, their application to complex fungal pathogens has remained limited due to the frequent inability to culture these pathogens in the absence of their host and their large genome sizes. RESULTS: Here, we describe the development of Mobile And Real-time PLant disEase (MARPLE) diagnostics, a portable, genomics-based, point-of-care approach specifically tailored to identify individual strains of complex fungal plant pathogens. We used targeted sequencing to overcome limitations associated with the size of fungal genomes and their often obligately biotrophic nature. Focusing on the wheat yellow rust pathogen, Puccinia striiformis f.sp. tritici (Pst), we demonstrate that our approach can be used to rapidly define individual strains, assign strains to distinct genetic lineages that have been shown to correlate tightly with their virulence profiles and monitor genes of importance. CONCLUSIONS: MARPLE diagnostics enables rapid identification of individual pathogen strains and has the potential to monitor those with specific properties such as fungicide resistance directly from field-collected infected plant tissue in situ. Generating results within 48 h of field sampling, this new strategy has far-reaching implications for tracking plant health threats.

Differentiation of Pythium spp. from vegetable crops with molecular markers and sensitivity to azoxystrobin and mefenoxam.

BACKGROUND: Pythium species attack various vegetable crops causing seed, stem and root rot, and 'damping-off' after germination. Pythium diseases are prevalently controlled by two classes of fungicides, QoIs with azoxystrobin and phenlyamides with mefenoxam as representatives. The present study aimed to test the sensitivity of six Pythium species from different vegetable crops to azoxystrobin and mefenoxam and differentiating species based on ITS, cytochrome b and RNA polymerase I gene sequences. RESULTS: The inter- and intra-species sensitivity to azoxystrobin was found to be stable, with the exception of one Pythium paroecandrum isolate, which showed reduced sensitivity and two cytochrome b amino acid changes. For mefenoxam, the inter-species sensitivity was quite variable and many resistant isolates were found in all six Pythium species, but no RNA polymerase I amino acid changes were observed in them. ITS and cytochrome b phylogenetic analyses permitted a clear separation of Pythium species corresponding to globose- and filamentous-sporangia clusters. CONCLUSION: The results document the necessity of well-defined chemical control strategies adapted to different Pythium species. Since the intrinsic activity of azoxystrobin among species was stable and no resistant isolates were found, it may be applied without species differentiation, provided it is used preventatively to also control highly aggressive isolates. For a reliable use of mefenoxam, precise identification and sensitivity tests of Pythium species are crucial because its intrinsic activity is variable and resistant isolates may exist. Appropriate mixtures and/or alternation of products may help to further delay resistance development. © 2018 Society of Chemical Industry.

Identification of Tomato Disease Types and Detection of Infected Areas Based on Deep Convolutional Neural Networks and Object Detection Techniques.

This study develops tomato disease detection methods based on deep convolutional neural networks and object detection models. Two different models, Faster R-CNN and Mask R-CNN, are used in these methods, where Faster R-CNN is used to identify the types of tomato diseases and Mask R-CNN is used to detect and segment the locations and shapes of the infected areas. To select the model that best fits the tomato disease detection task, four different deep convolutional neural networks are combined with the two object detection models. Data are collected from the Internet and the dataset is divided into a training set, a validation set, and a test set used in the experiments. The experimental results show that the proposed models can accurately and quickly identify the eleven tomato disease types and segment the locations and shapes of the infected areas.

Color disease leaf image segmentation using NAMS superpixel algorithm.

BACKGROUND: Disease leaf segmentation in color image is used to explore the disease shape and lesion regions. It is of great significance for pathological diagnosis and pathological research. OBJECTIVE: This paper proposes a superpixel algorithm using Non-symmetry and Anti-packing Model with Squares (NAMS) for color image segmentation of leaf disease. METHODS: First of all, the NAMS model is presented for color leaf disease image representation. The model can segment images asymmetrically and preserve the characteristics of image context. Second, NAMS based superpixel (NAMS superpixel) algorithm is proposed for clustering pixels, which can represent large homogeneous areas by super squares. By this way, the impact of complex background and the data redundancy in image segmentation can be reduced. RESULTS: Experimental results indicate that compared with segmenting the original image directly and manipulating by Simple Linear Iterative Clustering (SLIC) superpixel, the proposed NAMS superpixel performs more excellently in not only saving storage but also adhering to the lesion region edge. CONCLUSIONS: The outcome of NAMS superpixel can be regarded as a preprocess procedure for leaf disease region detection since the method can segment the image into superpixel blocks and preserve the lesion area.

An explainable deep machine vision framework for plant stress phenotyping.

Current approaches for accurate identification, classification, and quantification of biotic and abiotic stresses in crop research and production are predominantly visual and require specialized training. However, such techniques are hindered by subjectivity resulting from inter- and intrarater cognitive variability. This translates to erroneous decisions and a significant waste of resources. Here, we demonstrate a machine learning framework's ability to identify and classify a diverse set of foliar stresses in soybean [Glycine max (L.) Merr.] with remarkable accuracy. We also present an explanation mechanism, using the top-K high-resolution feature maps that isolate the visual symptoms used to make predictions. This unsupervised identification of visual symptoms provides a quantitative measure of stress severity, allowing for identification (type of foliar stress), classification (low, medium, or high stress), and quantification (stress severity) in a single framework without detailed symptom annotation by experts. We reliably identified and classified several biotic (bacterial and fungal diseases) and abiotic (chemical injury and nutrient deficiency) stresses by learning from over 25,000 images. The learned model is robust to input image perturbations, demonstrating viability for high-throughput deployment. We also noticed that the learned model appears to be agnostic to species, seemingly demonstrating an ability of transfer learning. The availability of an explainable model that can consistently, rapidly, and accurately identify and quantify foliar stresses would have significant implications in scientific research, plant breeding, and crop production. The trained model could be deployed in mobile platforms (e.g., unmanned air vehicles and automated ground scouts) for rapid, large-scale scouting or as a mobile application for real-time detection of stress by farmers and researchers.

Use of Comparative Genomics-Based Markers for Discrimination of Host Specificity in Fusarium oxysporum.

The polyphyletic nature of many formae speciales of Fusarium oxysporum prevents molecular identification of newly encountered strains based on conserved, vertically inherited genes. Alternative molecular detection methods that could replace labor- and time-intensive disease assays are therefore highly desired. Effectors are functional elements in the pathogen-host interaction and have been found to show very limited sequence diversity between strains of the same forma specialis, which makes them potential markers for host-specific pathogenicity. We therefore compared candidate effector genes extracted from 60 existing and 22 newly generated genome assemblies, specifically targeting strains affecting cucurbit plant species. Based on these candidate effector genes, a total of 18 PCR primer pairs were designed to discriminate between each of the seven Cucurbitaceae-affecting formae speciales When tested on a collection of strains encompassing different clonal lineages of these formae speciales, nonpathogenic strains, and strains of other formae speciales, they allowed clear recognition of the host range of each evaluated strain. Within Fusarium oxysporum f. sp. melonis more genetic variability exists than anticipated, resulting in three F. oxysporum f. sp. melonis marker patterns that partially overlapped with the cucurbit-infecting Fusarium oxysporum f. sp. cucumerinum, Fusarium oxysporum f. sp. niveum, Fusarium oxysporum f. sp. momordicae, and/or Fusarium oxysporum f. sp. lagenariae For F. oxysporum f. sp. niveum, a multiplex TaqMan assay was evaluated and was shown to allow quantitative and specific detection of template DNA quantities as low as 2.5 pg. These results provide ready-to-use marker sequences for the mentioned F. oxysporum pathogens. Additionally, the method can be applied to find markers distinguishing other host-specific forms of F. oxysporumIMPORTANCE Pathogenic strains of Fusarium oxysporum are differentiated into formae speciales based on their host range, which is normally restricted to only one or a few plant species. However, horizontal gene transfer between strains in the species complex has resulted in a polyphyletic origin of host specificity in many of these formae speciales This hinders accurate and rapid pathogen detection through molecular methods. In our research, we compared the genomes of 88 strains of F. oxysporum with each other, specifically targeting virulence-related genes that are typically highly similar within each forma specialis Using this approach, we identified marker sequences that allow the discrimination of F. oxysporum strains affecting various cucurbit plant species through different PCR-based methods.

Automatic Image-Based Plant Disease Severity Estimation Using Deep Learning.

Automatic and accurate estimation of disease severity is essential for food security, disease management, and yield loss prediction. Deep learning, the latest breakthrough in computer vision, is promising for fine-grained disease severity classification, as the method avoids the labor-intensive feature engineering and threshold-based segmentation. Using the apple black rot images in the PlantVillage dataset, which are further annotated by botanists with four severity stages as ground truth, a series of deep convolutional neural networks are trained to diagnose the severity of the disease. The performances of shallow networks trained from scratch and deep models fine-tuned by transfer learning are evaluated systemically in this paper. The best model is the deep VGG16 model trained with transfer learning, which yields an overall accuracy of 90.4% on the hold-out test set. The proposed deep learning model may have great potential in disease control for modern agriculture.

Characterization of a soybean mosaic virus variant causing different diseases in Glycine max and Nicotiana benthamiana.

We discovered a soybean mosaic virus (SMV) variant (4278-1) that caused systemic infections in Nicotiana benthamiana plants, resulting in stem stunting and leaf shriveling. The virus had a particle morphology and incubation period similar to those of other SMV isolates but differed from them in the leaf symptoms it caused when infecting soybean and N. benthamiana. The genome of this variant consisted of a 9994-nt single-stranded RNA, which was different from most of the other known SMV isolates (approximately 9600 nt). Interestingly, we found evidence that two recombination events (nt 1-476 and nt 1145-1349) had occurred between 4278-1 and a watermelon mosaic virus analogue (WMV analogue), in the 5' untranslated region and the P1 cistron.

Plant Disease Severity Assessment-How Rater Bias, Assessment Method, and Experimental Design Affect Hypothesis Testing and Resource Use Efficiency.

The effect of rater bias and assessment method on hypothesis testing was studied for representative experimental designs for plant disease assessment using balanced and unbalanced data sets. Data sets with the same number of replicate estimates for each of two treatments are termed "balanced" and those with unequal numbers of replicate estimates are termed "unbalanced". The three assessment methods considered were nearest percent estimates (NPEs), an amended 10% incremental scale, and the Horsfall-Barratt (H-B) scale. Estimates of severity of Septoria leaf blotch on leaves of winter wheat were used to develop distributions for a simulation model. The experimental designs are presented here in the context of simulation experiments which consider the optimal design for the number of specimens (individual units sampled) and the number of replicate estimates per specimen for a fixed total number of observations (total sample size for the treatments being compared). The criterion used to gauge each method was the power of the hypothesis test. As expected, at a given fixed number of observations, the balanced experimental designs invariably resulted in a higher power compared with the unbalanced designs at different disease severity means, mean differences, and variances. Based on these results, with unbiased estimates using NPE, the recommended number of replicate estimates taken per specimen is 2 (from a sample of specimens of at least 30), because this conserves resources. Furthermore, for biased estimates, an apparent difference in the power of the hypothesis test was observed between assessment methods and between experimental designs. Results indicated that, regardless of experimental design or rater bias, an amended 10% incremental scale has slightly less power compared with NPEs, and that the H-B scale is more likely than the others to cause a type II error. These results suggest that choice of assessment method, optimizing sample number and number of replicate estimates, and using a balanced experimental design are important criteria to consider to maximize the power of hypothesis tests for comparing treatments using disease severity estimates.

Deep Neural Networks Based Recognition of Plant Diseases by Leaf Image Classification.

The latest generation of convolutional neural networks (CNNs) has achieved impressive results in the field of image classification. This paper is concerned with a new approach to the development of plant disease recognition model, based on leaf image classification, by the use of deep convolutional networks. Novel way of training and the methodology used facilitate a quick and easy system implementation in practice. The developed model is able to recognize 13 different types of plant diseases out of healthy leaves, with the ability to distinguish plant leaves from their surroundings. According to our knowledge, this method for plant disease recognition has been proposed for the first time. All essential steps required for implementing this disease recognition model are fully described throughout the paper, starting from gathering images in order to create a database, assessed by agricultural experts. Caffe, a deep learning framework developed by Berkley Vision and Learning Centre, was used to perform the deep CNN training. The experimental results on the developed model achieved precision between 91% and 98%, for separate class tests, on average 96.3%.