Vis-NIR Spectroscopy and Machine Learning Methods for the Discrimination of Transgenic Brassica napus L. and Their Hybrids with B. juncea

Abstract

The rapid advancement of genetically modified (GM) technology over the years has raised concerns about the safety of GM crops and foods for human health and the environment. Gene flow from GM crops may be a threat to the environment. Therefore, it is critical to develop reliable, rapid, and low-cost technologies for detecting and monitoring the presence of GM crops and crop products. Here, we used visible near-infrared (Vis-NIR) spectroscopy to distinguish between GM and non-GM Brassica napus, B. juncea, and F1 hybrids (B. juncea X GM B. napus). The Vis-NIR spectra were preprocessed with different preprocessing methods, namely normalization, standard normal variate, and Savitzky–Golay. Both raw and preprocessed spectra were used in combination with eight different chemometric methods for the effective discrimination of GM and non-GM plants. The standard normal variate and support vector machine combination was determined to be the most accurate model in the discrimination of GM, non-GM, and hybrid plants among the many combinations (99.4%). The use of deep learning in combination with Savitzky–Golay resulted in 99.1% classification accuracy. According to the findings, it is concluded that handheld Vis-NIR spectroscopy combined with chemometric analyses could be used to distinguish between GM and non-GM B. napus, B. juncea, and F1 hybrids.

Discrimination of Brassica juncea Varieties Using Visible Near-Infrared (Vis-NIR) Spectroscopy and Chemometrics Methods

“…From 650 to 750 nm, there was a sharp increase in the peak that remained higher absorbance value; later, there are no variations in remaining wavelength until 1200 nm. These results concurred with our previous research on the discrimination of B. napus and B. juncea using Vis-NIR spectroscopy [ 23 ]. The spectra were preprocessed to reduce systemic noise and emphasize differences between samples.…”

“…The spectra were preprocessed to reduce systemic noise and emphasize differences between samples. Using a number of preprocessing methods simultaneously will help us obtain a greater degree of classification accuracy and will allow us to select the best preprocessing approach for each sample [ 23 , 24 ]. It is difficult to discriminate the plant varieties only with the spectra shown in Figure 1 .…”

“…Similar results were witnessed by the use of Vis-NIR spectroscopy in the discrimination of Amaranthus sp. [ 11 ] and hybrids between B. napus and B. juncea [ 23 ].…”

“…Background signals arose in the raw spectra of samples due to system settings and external noise. As a result, numerous preprocessing procedures, such as normalization (area), standard normal variate (SNV), and derivatives (Savitzky–Golay with first differentiation), were used to reduce spectral noise and improve the accuracy of modeling approaches [ 20 , 23 ]. The efficiency of preprocessing methods was evaluated in comparison to raw spectra.…”

“…The modeling was performed with RapidMiner studios Version 9.0.002 (RapidMiner, Inc., Boston, MA, USA). Deep learning, decision trees, support vector machines (SVM), random forests, generalized linear model, rapid large margin, Naïve Bayes, and linear discriminant analysis were used in this study to find the best modeling technique with the highest classification accuracy [ 20 , 23 ]. The Aquap2 package created by Pollner and Kovacs [ 33 ] was also utilized in R-studio to apply the various preprocessing approaches and perform linear discriminant analysis.…”