Image-based assessment of plant disease progression identifies new genetic loci for resistance to Ralstonia solanacearum in tomato.

A major challenge in global crop production is mitigating yield loss due to plant diseases. One of the best strategies to control these losses is through breeding for disease resistance. One barrier to the identification of resistance genes is the quantification of disease severity, which is typically a subjective score from a human observer. We hypothesized that image-based, non-destructive measurements of plant morphology over an extended period after pathogen infection would capture subtle quantitative differences between genotypes, and thus enable identification of new disease resistance loci. To test this, we inoculated a genetically diverse biparental mapping population of Solanum lycopersium (tomato) with Ralstonia solanacearum, a soilborne pathogen that causes bacterial wilt disease. We acquired over 40,000 time-series images of disease progression in this population, and developed an image analysis pipeline providing a suite of 10 traits to quantify bacterial wilt disease based on plant shape and size. Quantitative trait loci (QTL) analyses using image-based phenotyping of single and multi-trait analyses identified QTL that were both unique and shared compared with those identified by human assessment of wilting, and could detect QTL earlier than human assessment. Expanding the phenotypic space of disease with image-based, non-destructive phenotyping allowed both earlier detection and identified new genetic components of resistance.