Identifying Optimal Wavelengths as Disease Signatures Using Hyperspectral Sensor and Machine Learning

Hyperspectral sensors combined with machine learning are increasingly utilized in agricultural crop systems for diverse applications, including plant disease detection. This study was designed to identify the most important wavelengths to discriminate between healthy and diseased peanut (Arachis hypogaea L.) plants infected with Athelia rolfsii, the causal agent of peanut stem rot, using in-situ spectroscopy and machine learning. In greenhouse experiments, daily measurements were conducted to inspect disease symptoms visually and to collect spectral reflectance of peanut leaves on lateral stems of plants mock-inoculated and inoculated with A. rolfsii. Spectrum files were categorized into five classes based on foliar wilting symptoms. Five feature selection methods were compared to select the top 10 ranked wavelengths with and without a custom minimum distance of 20 nm. Recursive feature elimination methods outperformed the chi-square and SelectFromModel methods. Adding the minimum distance of 20 nm into the top selected wavelengths improved classification performance. Wavelengths of 501–505, 690–694, 763 and 884 nm were repeatedly selected by two or more feature selection methods. These selected wavelengths can be applied in designing optical sensors for automated stem rot detection in peanut fields. The machine-learning-based methodology can be adapted to identify spectral signatures of disease in other plant-pathogen systems.

Proteometabolomic Study of Compatible Interaction in Tomato Fruit Challenged with Sclerotinia rolfsii Illustrates Novel Protein Network during Disease Progression

Fruit is an assimilator of metabolites, nutrients, and signaling molecules, thus considered as potential target for pathogen attack. In response to patho-stress, such as fungal invasion, plants reorganize their proteome, and reconfigure their physiology in the infected organ. This remodeling is coordinated by a poorly understood signal transduction network, hormonal cascades, and metabolite reallocation. The aim of the study was to explore organ-based proteomic alterations in the susceptibility of heterotrophic fruit to necrotrophic fungal attack. We conducted time-series protein profiling of Sclerotinia rolfsii invaded tomato (Solanum lycopersicum) fruit. The differential display of proteome revealed 216 patho-stress responsive proteins (PSRPs) that change their abundance by more than 2.5-fold. Mass spectrometric analyses led to the identification of 56 PSRPs presumably involved in disease progression; regulating diverse functions viz. metabolism, signaling, redox homeostasis, transport, stress-response, protein folding, modification and degradation, development. Metabolome study indicated differential regulation of organic acid, amino acids, and carbohydrates paralleling with the proteomics analysis. Further, we interrogated the proteome data using network analysis that identified two significant functional protein hubs centered around malate dehydrogenase, T-complex protein 1 subunit gamma, and ATP synthase beta. This study reports, for the first-time, kinetically controlled patho-stress responsive protein network during post-harvest storage in a sink tissue, particularly fruit and constitute the basis toward understanding the onset and context of disease signaling and metabolic pathway alterations. The network representation may facilitate the prioritization of candidate proteins for quality improvement in storage organ.

Mycelial Compatibility and Pathogenic Diversity Among Sclerotium rolfsii Isolates in the Southern United States

Sclerotium rolfsii is a soilborne fungus that causes southern blight on a wide range of plants in tropical and subtropical regions of the world. Eighty-four isolates collected from Florida, Georgia, Louisiana, South Carolina, Texas, and Virginia were paired and assigned to 23 mycelial compatibility groups (MCGs), of which 11 MCGs consisted of a single isolate. Isolates within an MCG typically originated from different hosts and different geographical areas, with the exception of MCG 11. In all, 13 of the 15 isolates in MCG 11 originated from peanut in Georgia and Florida, while the other 2 isolates originated from potato in Virginia and from the ornamental Barlaeria cristata in Florida. Significant differences in the size and number of sclerotia produced in vitro existed between isolates from peanut and other hosts. Nineteen isolates representative of the most common MCGs (1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 13, 14, 15, 16, 17, and 18) were tested for pathogenicity on tomato, pepper, and peanut. All isolates were pathogenic on all hosts but virulence differed significantly among isolates. Isolates collected from peanut were the most virulent on all three hosts compared with isolates collected from tomato and pepper. 'Georgia Green' peanut was more susceptible to peanut isolates from Georgia than to the other tested isolates. Of the two tomato entries, the commercial tomato 'Tygress' was less susceptible than the previously reported resistant breeding line 5635M to many of the S. rolfsii isolates tested, with the exception of the peanut isolates collected from Georgia. These initial findings suggest that considerable variation exists among S. rolfsii isolates throughout the southern United States, with some indications of specialization for the isolates collected from peanut.

Phylogenetic placement of plant pathogenic Sclerotium species among teleomorph genera

Phylogenetic analyses and morphological characteristics were used to assess the taxonomic placement of eight plant-pathogenic Sclerotium species. Members of this genus produce only sclerotia and no fruiting bodies or spores, so Sclerotium species have been difficult to place taxonomically. Sequences of rDNA large subunit (LSU) and internal transcribed spacer (ITS) regions were determined for isolates of Sclerotium cepivorum, S. coffeicola, S. denigrans, S. hydrophilum, Ceratorhiza oryzae-sativae, S. perniciosum, S. rhizodes, S. rolfsii and S. rolfsii var. delphinii. Parsimony analysis grouped two species previously thought to be in the Basidiomycota, S. denigrans and S. perniciosum, within the Ascomycota; these species were found to have affinities with the teleomorph genera Sclerotinia and Stromatinia and the asexual Sclerotium cepivorum, which was known earlier to be related to Sclerotinia species. The other Sclerotium species were placed in one of two basidiomycetous groups, genera Athelia or Ceratobasidium. Based on rDNA analysis and morphology the basidiomycetous Sclerotium hydrophilum and S. rhizodes were transferred to genus Ceratorhiza, the anamorph of Ceratobasidium species. Sclerotium coffeicola was found to be close to S. rolfsii var. delphinii and S. rolfsii var. rolfsii, which was shown earlier to have an Athelia teleomorph.

Overwintering of Sclerotium rolfsii and S. rolfsii var. delphinii in Different Latitudes of the United States

Previously known only from the southern United States, hosta petiole rot recently appeared in the northern United States. Sclerotium rolfsii var. delphinii is believed to be the predominant petiole rot pathogen in the northern United States, whereas S. rolfsii is most prevalent in the southern United States. In order to test the hypothesis that different tolerance to climate extremes affects the geographic distribution of these fungi, the survival of S. rolfsii and S. rolfsii var. delphinii in the northern and southeastern United States was investigated. At each of four locations, nylon screen bags containing sclerotia were placed on the surface of bare soil and at 20-cm depth. Sclerotia were recovered six times from November 2005 to July 2006 in North Dakota and Iowa, and from December 2005 to August 2006 in North Carolina and Georgia. Survival was estimated by quantifying percentage of sclerotium survival on carrot agar. Sclerotia of S. rolfsii var. delphinii survived until at least late July in all four states. In contrast, no S. rolfsii sclerotia survived until June in North Dakota or Iowa, whereas 18.5% survived until August in North Carolina and 10.3% survived in Georgia. The results suggest that inability to tolerate low temperature extremes limits the northern range of S. rolfsii.