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.

Routine strain-level identification of plant pathogens directly from symptomatic tissue could significantly improve plant disease control and prevention. Here we tested the Oxford Nanopore Technologies (ONT) MinION sequencer for metagenomic sequencing of tomato plants either artificially inoculated with a known strain of the bacterial speck pathogen Pseudomonas syringae pv. tomato or collected in the field and showing bacterial spot symptoms caused by one of four Xanthomonas species. After species-level identification via ONT’s WIMP software and the third-party tools Sourmash and MetaMaps, we used Sourmash and MetaMaps with a custom database of representative genomes of bacterial tomato pathogens to attempt strain-level identification. In parallel, each metagenome was assembled and the longest contigs were used as query with the genome-based microbial identification Web service LINbase. Both the read-based and assembly-based approaches correctly identified P. syringae pv. tomato strain T1 in the artificially inoculated samples. The pathogen strain in most field samples was identified as a member of Xanthomonas perforans group 2. This result was confirmed by whole genome sequencing of colonies isolated from one of the samples. Although in our case metagenome-based pathogen identification at the strain level was achieved, caution still must be exercised in interpreting strain-level results because of the challenges inherent to assigning reads to specific strains and the error rate of nanopore sequencing.

Global travel and trade in combination with climate change are expanding the geographic distribution of plant pathogens. The bacterium Xylella fastidiosa is a prime example. Native to the Americas, it has spread to Europe, Asia, and the Middle East. To assess the risk that pathogen introductions pose to crops in newly invaded areas, it is key to survey their diversity, host range, and disease incidence in relation to climatic conditions where they are already present. We performed a survey of X. fastidiosa in grapevine in Virginia using a combination of quantitative PCR, multilocus sequencing, and metagenomics. We also analyzed samples from deciduous trees with leaf scorch symptoms. X. fastidiosa subspecies fastidiosa was identified in grapevines in all regions of the state, even in Northern Virginia, where the temperature was below −9°C for 10 days per year on average in the years preceding sampling. Unexpectedly, we also found for the first time grapevine samples infected with X. fastidiosa subspecies multiplex ( Xfm). The Xfm lineage found in grapevines had been previously isolated from blueberries in the Southeastern United States and was distinct from that found in deciduous trees in Virginia. The obtained results will be important for risk assessment of X. fastidiosa introductions in other parts of the world.