Comparing Characterization Methods of Fusarium Ear Rot Resistance in Corn
Is Version Of
Fusarium Ear Rot in corn is a disease that produces toxins known as fumonisins that are harmful to humans and other animals. Characterizing and improving the resistance to Fusarium Ear Rot in corn is important to corn seed producers like Corteva to reduce fumonisin exposure in humans and livestock. To characterize corn hybrids Corteva visually inspects ears each from a single corn variety and scores them based on observed symptoms. This experiment was intended to compare different methods of characterizing Fusarium Ear Rot resistance in order to improve the accuracy and efficiency of Corteva’s characterization methods. Four methods of characterization were compared. The first method was the traditional visual assessment. The second method was quantitative analysis of a grain sample for the amount of fumonisin toxins present. The third method was collecting images of ear piles and using a computer program to quantify the amount of symptomatic corn versus healthy corn based on color. The fourth method was collecting images of loose kernels and using a computer program to measure the amount of symptomatic corn versus healthy corn based on color. The third and fourth method use digital image analysis which is known as photometry. Statistical analysis of the data included Pearson’s correlation, linear regression and logistic regression analyses to compare the different methods. Correlation and regression statistics indicate a strong relationship between the four methods of characterization. Whole ear photometry performed equally as well as visual inspection at predicting the amount of fumonisin toxins present in a sample. Loose kernel photometry did not perform as well as the visual method or whole ear photometry, but still performed well enough at predicting fumonisin concentrations in a sample to potentially be used for characterization. Photometry shows potential as an effective characterization method for Fusarium Ear Rot resistance that could reduce the potential for human error and be more easily automated than visual inspections or laboratory analysis for toxins.