Discrimination of genetically modified sugar beets based on terahertz spectroscopy

•A deep investigation into the THz spectra of GM and non-GM sugar beets•Three chemometrics methods, PCA, DA and DPLS, are used to discriminate GM and non-GM sugar beets.•The proposed method is simple, rapid, non-destructive, and requires no pretreatments.

The objective of this paper was to apply terahertz (THz) spectroscopy combined with chemometrics techniques for discrimination of genetically modified (GM) and non-GM sugar beets. In this paper, the THz spectra of 84 sugar beet samples (36 GM sugar beets and 48 non-GM ones) were obtained by using terahertz time-domain spectroscopy (THz–TDS) system in the frequency range from 0.2 to 1.2 THz. Three chemometrics methods, principal component analysis (PCA), discriminant analysis (DA) and discriminant partial least squares (DPLS), were employed to classify sugar beet samples into two groups: genetically modified organisms (GMOs) and non-GMOs. The DPLS method yielded the best classification result, and the percentages of successful classification for GM and non-GM sugar beets were both 100%. Results of the present study demonstrate the usefulness of THz spectroscopy together with chemometrics methods as a powerful tool to distinguish GM and non-GM sugar beets.

Graphical abstractFigure optionsDownload full-size imageDownload as PowerPoint slide