High-throughput phenotyping of leaf discs infected with grapevine downy mildew using shallow convolutional neural networks

Abstract

Objective and standardized recording of disease severity in mapping crosses and breeding lines is a crucial step in characterizing resistance traits utilized in breeding programs and to conduct QTL or GWAS studies. Here we report a system for automated high-throughput scoring of disease severity on inoculated leaf discs. As proof of concept, we used leaf discs inoculated with Plasmopara viticola causing grapevine downy mildew (DM). This oomycete is one of the major grapevine pathogens and has the potential to reduce grape yield dramatically if environmental conditions are favorable. Breeding of DM resistant grapevine cultivars is an approach for a novel and more sustainable viticulture. This involves the evaluation of several thousand inoculated leaf discs from mapping crosses and breeding lines every year. Therefore, we trained a shallow convolutional neural-network (SCNN) for efficient detection of leaf disc segments showing P. viticola sporangiophores. We could illustrate a high and significant correlation with manually scored disease severity used as ground truth data for evaluation of the SCNN performance. Combined with an automated imaging system, this leaf disc-scoring pipeline has the potential to reduce the amount of time during leaf disc phenotyping considerably. The pipeline with all necessary documentation for adaptation to other pathogens is freely available.

Phenotyping strategies for Elsinöe ampelina symptoms in grapevine (Vitis spp.)

“…During anthracnose evaluation in this study, the classification of progenies into groups (score and SAD) showed less variation between the experiments than imaging analysis (number of spot and severity). Imaging analysis and automatic phenotyping increase the precision during the disease evaluation because measure real values (Pujari et al, 2015; Zendler et al, 2021). However, grouping into classes or scores is a reliable assessment because the range of each group/score classify the genotypes from high resistance to susceptible, resulting in less experimental variation during different experiments than actual value (counted) (Murria et al, 2018; Poolsawat et al, 2012; Rex et al, 2014).…”

“…However, there is no study about young plant be more susceptible than the older ones that can result in difference during the phenotyping in different crop seasons. For other grapevine diseases, imaging analysis was used to phenotype disease symptoms, but the real severity or pathogen sporulation was grouped into classes of resistance/susceptibility, from extremely resistance to extremely sensitive against the pathogens, as a measure to minimize the data variation (Özer et al, 2021; Rex et al, 2014; Zendler et al, 2021).…”