Muthreich Florian, Tafintseva Valeria, Zimmermann Boris, Kohler Achim, Vila-Viçosa Carlos M, Seddon Alistair W R
University of Bergen, Department of Biological Sciences, Bergen, Norway.
Norwegian University of Life Sciences, Faculty of Science and Technology, Ås, Norway.
Appl Spectrosc. 2025 Jul;79(7):1142-1154. doi: 10.1177/00037028251334405. Epub 2025 May 23.
Vibrational spectroscopy is gaining popularity for understanding ecological and evolutionary patterns in plants, particularly in relation to the analysis of pollen grains. So far, Fourier transform infrared spectroscopy (FT-IR) has been the main approach used to classify pollen grains based on chemical variations. However, FT-IR may be less suitable for detecting differences in the pollen grain exine, mainly composed of sporopollenin. In contrast, Raman spectroscopy has increased sensitivity for the main chemical components found within sporopollenins. We compare the classification performance and chemical information provided by FT-IR and FT-Raman using a large dataset of L. pollen, comprising five species in three sections: (i) : . , (ii) : , , and (iii) : , ). Here, we used multiblock sparse partial least squares discriminant analyses (MB-sPLS-DA) analyses to directly compare the two infrared methods. Both FT-IR and FT-Raman successfully classified pollen to section level (100% accuracy). At the species level our models achieved ∼90% accuracy for FT-Raman and FT-IR separately and in the combined multiblock model. The multiblock results showed an increased number of sporopollenin peaks observed in FT-Raman spectra as compared to FT-IR. These peaks are also of a higher importance for classification. Results also showed differences in the types of vibrations that are of diagnostic value for the two infrared methods. CH deformations are more important in FT-Raman, while C-O-C, C-O, and C = O stretches are more important for FT-IR-based identification of pollen. These vibrations are indicators of carbohydrates, proteins and lipids. FT-Raman provides equally successful diagnostic potential to FT-IR, but uses more chemical information based on variations in sporopollenin chemistry than FT-IR. We suggest that the combined analysis of FT-IR and FT-Raman using multiblock analysis has great potential for classification.
振动光谱法在理解植物的生态和进化模式方面越来越受欢迎，特别是在花粉粒分析方面。到目前为止，傅里叶变换红外光谱（FT-IR）一直是基于化学变化对花粉粒进行分类的主要方法。然而，FT-IR可能不太适合检测主要由孢粉素组成的花粉粒外壁的差异。相比之下，拉曼光谱对孢粉素中发现的主要化学成分具有更高的灵敏度。我们使用包含三个组中五个物种的大丽花花粉的大型数据集，比较了FT-IR和傅里叶变换拉曼光谱（FT-拉曼）提供的分类性能和化学信息：（i）：......，（ii）：......，......，以及（iii）：......，......）。在这里，我们使用多块稀疏偏最小二乘判别分析（MB-sPLS-DA）分析来直接比较这两种红外方法。FT-IR和FT-拉曼都成功地将花粉分类到组水平（准确率100%）。在物种水平上，我们的模型分别对FT-拉曼和FT-IR以及组合的多块模型实现了约90%的准确率。多块结果表明，与FT-IR相比，FT-拉曼光谱中观察到的孢粉素峰数量增加。这些峰对分类也更重要。结果还显示了两种红外方法具有诊断价值的振动类型的差异。CH变形在FT-拉曼中更重要，而C-O-C、C-O和C = O伸缩对基于FT-IR的花粉鉴定更重要。这些振动是碳水化合物、蛋白质和脂质的指标。FT-拉曼提供了与FT-IR同样成功的诊断潜力，但比FT-IR使用了更多基于孢粉素化学变化的化学信息。我们建议使用多块分析对FT-IR和FT-拉曼进行联合分析在分类方面具有很大潜力。
