Imaging for improving Fusarium-damaged kernel and deoxynivalenol resistance in Canadian wheat


Term
2023 - 2026
Sask Wheat Funding
$18,779
Status
status in progress

Lead Researcher

Lead Researcher

Dr. Samia BerraiesAgriculture and Agri-Food Canada (AAFC) Swift Current
Dr. Samia Berraies

Funding Partners: 

Saskatchewan Ministry of Agriculture (SMA) – Agriculture Development Fund (ADF), Alberta Grains, Western Grains Research Foundation (WGRF), Manitoba Crop Alliance (MCA)


Summary

The project aims to increase the resolution for Fusarium damaged kernel (FDK) assessment with RGB colour based imaging.

Objectives: 

  1. A classification model of Fusarium damaged kernels (FDK) based on the severity of infection will be developed using the VibeQM3 instrument, a high throughput image analyzer to improve resolution of FDK resistance. Infected durum and hexaploid wheat samples with varying levels of FHB resistance collected from several Fusarium head blight (FHB) nurseries will be assessed using the image analyzer. The generated images will be analyzed to distinguish between different kernel infection levels based on a classification model with colour descriptors L (lightness), C (chroma), and H (Hue) in the LCH colour space; a* (red/green coordinate) and b* (yellow/blue coordinate) in the L*a*b* colour space and R (red), G (green), and B (blue) in the RGB colour space.
  2. To associate deoxynivalenol (DON) accumulation to the different classes of FDK and develop a statistical model to predict DON. Several durum and hexaploid wheat controls used in the wheat registration trial with a range of FDK severity spanning the classes identified in Objective 1 will be scanned using the VibeQM3 image analyzer to generate high resolution images. After generating the images, those samples will be evaluated for DON concentration using enzyme linked immunosorbent assay (ELISA) methodology. The generated DON data will be correlated to the image information associated with the different FDK classes and a statistical model will be developed to predict DON concentration based on FDK infection using R and/or python software.
  3. To validate QTL associated with FDK classes and DON resistance in hexaploid and durum genetic populations. Two doubled haploid (DH) genetic populations developed through previous projects will be evaluated in multiple FHB inoculated nurseries and processed after harvest for FDK classification using the VibeQM3. DON content associated with each FDK class will be afterwards predicted using the model developed in Objective 2. We will perform QTL analysis for FDK and DON using available 90K genotyping data. The resistance QTL associated with FDK classes and associated DON content generated with the image analyzer will be compared with the resistance QTL associated with FDK and DON previously generated through the standard protocol, consisting of visual assessment for FDK and ELISA methodology for DON determination.