大豆是一种重要的油料和提供蛋白的作物，过去的几十年，大豆基因改造取得了巨大进展。尽管在田间区分常规和转基因大豆种子以及杂交后代很困难，也要对基因漂移的可能性进行评估。我们通过结合化学计量法的多光谱成像系统，来检验无损鉴别区分常规与抗草甘膦的大豆种子的可行性。

   主成分分析法（PCA），偏较小二乘判别分析(PLSDA)，较小二乘支撑向量机 (LS-SVM)以及BP神经网络（(BPNN)用来对大豆种子进行区分。目前结果显示在常规和抗草甘膦大豆种子以及其后代之间有显著差异，很容易成像显示，鉴别率较高（BPNN法，98%）。结论表明多光谱成像系统结合化学计量法是有效鉴别转基因大豆种子的较具前景的技术。详细产品介绍请参见VideometerLab 产品链接。

Non-destructive discrimination of conventional and glyphosate-resistant soybean seeds and their hybrid descendants using multispectral imaging and chemometric methods

Soybean is an important oil- and protein-producing crop and over the last few decades soybean genetic transformation has made rapid strides. The probability of occurrence of transgene flow should be assessed, although the discrimination of conventional and transgenic soybean seeds and their hybrid descendants is difficult in fields. The feasibility of non-destructive discrimination of conventional and glyphosate-resistant soybean seeds and their hybrid descendants was examined by a multispectral imaging system combined with chemometric methods.

Principal component analysis (PCA), partial least squares discriminant analysis (PLSDA), least squares-support vector machines (LS-SVM) and back propagation neural network (BPNN) methods were applied to classify soybean seeds. The current results demonstrated that clear differences among conventional and glyphosate-resistant soybean seeds and their hybrid descendants could be easily visualized and an excellent classification (98% with BPNN model) could be achieved. It was concluded that multispectral imaging together with chemometric methods would be a promising technique to identify transgenic soybean seeds with high efficiency.