RUST: a Robust User-friendly Script Tool for rapid assessment of rust disease
Trabajo presentado en el III Spanish Symposium on Physiology and Breeding of Cereals (III Simposio Español de Fisiología y Mejora de Cereales, SEFiMeC), celebrado online el 17 y 18 de noviembre de 2020. ; During the last years a wealth of genomic data regarding many plant species have been generated taking advance of the decline of the cost of sequencing, creating many new opportunities. However, at the same time, phenotyping has become the main bottleneck in plant breeding and fundamental plant science. This is particularly true for the assessment of plant diseases where, in addition to the time consuming evaluations, we have often to deal with the subjectivity of macroscopic visual assessments. In this work, we have developed an open script for semi-automated evaluation of rust disease. The script function under the free image software Fiji (developed from Image J), which is a well-recognized software among scientific community. The script allow the counting of the number of pustules in leaf segments by a color transformation tool, and provides three different automation modes. The script opens images sequentially and records infection frequency (pustules per area) (semi-)automatically for high-throughput analysis. Furthermore, it can manage several scanned leaf segments in the same image, consecutively selecting the desired segments. The script has been validated with nearly 900 samples from 80 oat genotypes ranging from resistant to susceptible and from very light to heavily infected leaves showing a high accuracy with a Lin's concordance correlation coefficient of 0.99. The analysis show a high repeatability as indicated by the low variation coefficients obtained when repeating the measurement of the same samples. The script also has optional steps for calibration and training to ensure accuracy, even in low-resolution images. This script can evaluate efficiently hundreds of leaves facilitating the screening of novel sources of resistance to this important cereal disease. ; This work was supported by the Spanish Ministry of Science and Innovation [PID2019-104518RB-I00], and regional government through the AGR-253 group.