We have recently applied Raman spectroscopy for biochemical analysis of cell groups treated with 2.5, 5, 7.5 and 10 micromoles per liter of carboplatin and cisplatin, respectively, for apoptosis detection. The continuously decreased intensities of Raman bands assigned to the proteins and deoxyribonucleic acid can be observed as the concentrations of drugs increased, suggesting the cell damage induced by carboplatin and cisplatin. Principal component analysis combined with linear discriminate analysis and support vector machines analysis of the obtained spectral data was used to develop diagnostic algorithms. When cell groups were treated with the same concentration, both principal component analysis combined with linear discriminate analysis and support vector machines analysis executed between control group and groups treated with cisplatin, produce a lower error than those treated with carboplatin, suggesting the more cell damage induced by cisplatin. Among treated groups, an accuracy of 100% is only observed in the group treated with 10 micromoles per liter of cisplatin, demonstrating the apoptosis. Those Statistical analysis results are consistent with the biochemical and Raman spectral analysis. This study indicated that Raman spectroscopy in conjunction with principal component analysis combined with linear discriminate analysis and support vector machines analysis has great potential use in label-free detection of apoptosis in cells and label-free assessment of drug efficacy.

我们最近将拉曼光谱法用于分别用每升卡铂和顺铂2.5、5、7.5和10微摩尔处理的细胞组的生化分析，以检测细胞凋亡。随着药物浓度的增加，可以观察到分配给蛋白质和脱氧核糖核酸的拉曼谱带强度持续下降，这表明卡铂和顺铂诱导的细胞损伤。主成分分析结合线性判别分析和支持向量机对获得的光谱数据进行分析，以开发诊断算法。当以相同的浓度处理细胞组时，在对照组和顺铂处理组之间进行主成分分析，线性判别分析和支持向量机分析，与卡铂治疗相比，产生的错误更低，提示顺铂诱导的细胞损伤更大。在治疗组中，仅在每升顺铂10微摩尔的治疗组中观察到100％的准确性，这表明细胞凋亡。这些统计分析结果与生化和拉曼光谱分析一致。这项研究表明，拉曼光谱结合主成分分析，线性判别分析和支持向量机分析相结合，在无标记的细胞凋亡检测和无标记的药物疗效评估中具有巨大的潜力。仅在每升顺铂用10微摩尔化合物治疗的组中观察到100％的准确性，这表明细胞凋亡。这些统计分析结果与生化和拉曼光谱分析一致。这项研究表明，拉曼光谱结合主成分分析，线性判别分析和支持向量机分析相结合，在无标记的细胞凋亡检测和无标记的药物疗效评估中具有巨大的潜力。仅在每升顺铂用10微摩尔化合物治疗的组中观察到100％的准确性，这表明细胞凋亡。这些统计分析结果与生化和拉曼光谱分析一致。这项研究表明，拉曼光谱结合主成分分析，线性判别分析和支持向量机分析相结合，在无标记的细胞凋亡检测和无标记的药物疗效评估中具有巨大的潜力。