instaGraminoid, a Novel Colorimetric Method to Assess Herbicide Resistance, Identifies Patterns of Cross-Resistance in Annual Ryegrass

Abstract

<jats:p>Herbicide resistance in agricultural weeds is a global problem with an increasing understanding that it is caused by multiple genes leading to quantitative resistance. These quantitative patterns of resistance are not easy to decipher with mortality assays alone, and there is a need for straightforward and unbiased protocols to accurately assess quantitative herbicide resistance.<jats:italic> instaGraminoid</jats:italic>—a computer vision and statistical analysis package—was developed as an automated and scalable method for quantifying herbicide resistance. The package was tested in rigid ryegrass (<jats:italic>Lolium rigidum</jats:italic>), the most noxious and highly resistant weed in Australia and the Mediterranean region. This method provides quantitative measures of the degree of chlorosis and necrosis of individual plants which was shown to accurately reflect herbicide resistance. We were able to reliably characterise resistance to four herbicides with different sites of action (glyphosate, sulfometuron, terbuthylazine, and trifluralin) in two<jats:italic> L. rigidum</jats:italic> populations from Southeast Australia. Cross-validation of the method across populations and herbicide treatments showed high repeatability and transferability. Significant positive correlations in resistance of individual plants were observed across herbicides, which suggest either the accumulation of herbicide-specific resistance alleles in single genotypes (multiple stacked resistance) or the presence of general broad-effects resistance alleles (cross-resistance). We used these quantitative estimates of cross-resistance to simulate how resistance development under an herbicide rotation strategy is likely to be higher than expected.</jats:p>

Increasing the value and efficiency of herbicide resistance surveys

“…Reducing the assessment delay and reliance on visual estimates would improve repeatability and efficiency of the process. Computer vision tools and machine learning for image‐contextual understanding may offer opportunities to identify seedling chlorosis and necrosis 77 . On an individual plant assessment level, analysis of red‐green‐blue colour images of L. rigidum seedlings at 0, 7‐ and 14‐day intervals post‐treatment with glyphosate, sulfometuron, terbuthylazine and trifluralin, correctly classified 95% of the treated plants as either resistant or susceptible 77 .…”

“…Computer vision tools and machine learning for image‐contextual understanding may offer opportunities to identify seedling chlorosis and necrosis 77 . On an individual plant assessment level, analysis of red‐green‐blue colour images of L. rigidum seedlings at 0, 7‐ and 14‐day intervals post‐treatment with glyphosate, sulfometuron, terbuthylazine and trifluralin, correctly classified 95% of the treated plants as either resistant or susceptible 77 . When trained on glyphosate‐treated plants and tested on unseen glyphosate‐treated test sets, the model correctly classified 100% of plants.…”

Discrimination between protoporphyrinogen oxidase–inhibiting herbicide-resistant and herbicide-susceptible redroot pigweed (Amaranthus retroflexus) with spectral reflectance

“…Chloroplast integrity is compromised, and vegetative tissue loses green pigment (Carter 1993; Major et al 2003). Chlorophyll content and spectral reflectance are correlated; thus, PPO-inhibiting herbicide-treated plants could be imaged and analyzed to detect differences in spectral reflectance and thus confirm resistance (Gitelson et al 2003a; Major et al 2003; Paril and Fournier-Level 2019). Previous research has demonstrated that spectral reflectance can discriminate between herbicide-treated plants, weed species, and herbicide-resistant weed biotypes (Everman et al 2008; Reddy et al 2014; Sanders et al 2021; Zhao et al 2014).…”