Spectral signatures in the UV range can be combined with secondary plant metabolites by deep learning to characterize barley–powdery mildew interaction

In recent studies, the potential of hyperspectral sensors for analysis plant–pathogen interactions was expanded to ultraviolet range (UV; 200–380 nm) monitor stress processes in plants. A imaging set-up established highlight influence early on secondary plant metabolites. this study, three different barley lines inoculated with Blumeria graminis f. sp. hordei (Bgh, powdery mildew) were investigated. One susceptible genotype (cv. Ingrid, wild type) and two resistant genotypes (Pallas 01, Mla1- Mla12-based resistance Pallas 22, mlo5-based resistance) used. During first 5 days after inoculation (dai) reflectance patterns recorded metabolites relevant host–pathogen studied parallel. Hyperspectral measurements UV revealed that a differentiation between Bgh is possible, distinct each genotype. The extracted analysed pigments flavonoids correlated spectral data recorded. classification noninoculated samples deep learning high performance can be achieved self-attention networks. subsequent feature importance identified wavelengths as most important classification, these linked flavonoids. allows characterization reactions, changes metabolites, has advantage being non-invasive method. It therefore enables greater understanding reactions biotic stress, well reactions.