Image-based time series analysis to establish differential disease progression for two Fusarium head blight pathogens in oat spikelets with variable resistance

Abstract

Oat-based value-added products have increased their value as healthy foodstuff. Fusarium head blight (FHB) infections and the mycotoxins accumulated to the oat seeds, however, pose a challenge to oat production. The FHB infections are predicted to become more prevalent in the future changing climates and under more limited use of fungicides. Both these factors increase the pressure for breeding new resistant cultivars. Until now, however, genetic links in oats against FHB infection have been difficult to identify. Therefore, there is a great need for more effective breeding efforts, including improved phenotyping methods allowing time series analysis and the identification of molecular markers during disease progression. To these ends, dissected spikelets of several oat genotypes with different resistance profiles were studied by image-based methods during disease progression by Fusarium culmorum or F. langsethiae species. The chlorophyll fluorescence of each pixel in the spikelets was recorded after inoculation by the two Fusarium spp., and the progression of the infections was analyzed by calculating the mean maximum quantum yield of PSII (Fv/Fm) values for each spikelet. The recorded values were (i) the change in the photosynthetically active area of the spikelet as percentage of its initial size, and (ii) the mean of Fv/Fm values of all fluorescent pixels per spikelet post inoculation, both indicative of the progression of the FHB disease. The disease progression was successfully monitored, and different stages of the infection could be defined along the time series. The data also confirmed the differential rate of disease progression by the two FHB causal agents. In addition, oat varieties with variable responses to the infections were indicated.

The Combination of Low-Cost, Red–Green–Blue (RGB) Image Analysis and Machine Learning to Screen for Barley Plant Resistance to Net Blotch

“…Over the last decades, efforts have focused on the development of high-throughput phenotyping methods to overcome these constraints [24][25][26]; one potentially robust approach relies on using image analysis to measure and analyze plant health [27][28][29][30]. This simple and low-cost solution mostly uses red-green-blue (RGB) imagery to extract information about the shape, texture, and color of plants, and has shown promise for the evaluation of different abiotic and biotic stresses even before the typical symptoms manifest [31,32].…”