School of Traditional Chinese Medicine, Guangdong Pharmaceutical University, Guangzhou 510006, PR China; The Key Unit of Chinese Medicine Digitalization Quality Evaluation of SATCM, Guangzhou 510006, PR China; The Research Center for Quality Engineering Technology of Traditional Chinese Medicine in Guangdong Universities, Guangzhou 510006, PR China.
School of Traditional Chinese Medicine, Guangdong Pharmaceutical University, Guangzhou 510006, PR China; The Key Unit of Chinese Medicine Digitalization Quality Evaluation of SATCM, Guangzhou 510006, PR China; The Research Center for Quality Engineering Technology of Traditional Chinese Medicine in Guangdong Universities, Guangzhou 510006, PR China.
Spectrochim Acta A Mol Biomol Spectrosc. 2018 May 5;196:209-214. doi: 10.1016/j.saa.2018.02.021. Epub 2018 Feb 7.
To establish calibration models for simultaneous determination of contents of liquirtin and glycyrrhizic acid, and to investigate the variable selection methods.
The contents of liquirtin and glycyrrhizic acid determined by HPLC were as the reference values, which were associated with samples spectra by using near infrared spectrum (NIR) analysis technology. Calibration models were developed using partial least squares (PLS) regression algorithm, and evaluated by the independent dataset test with calculating the metrics of coefficients of determination of calibration and prediction (R, R), the root mean square errors of calibration and prediction (RMSEC, RMSEP), the mean absolute errors of calibration and prediction (MAEC, MAEP), and the residual prediction deviation (RPD). Five variable selection methods including variable importance in projection (VIP), competitive adaptive reweighted sampling (CARS), Monte Carlo uninformative variable elimination (MCUVE), particle swarm optimization (PSO) and genetic algorithm (GA), were investigated.
Compared to the original full spectra, both quantification models for liquirtin and glycyrrhizic acid performed better with a clear ranking of GA>PSO>CARS>MCUVE≅VIP>Full. Especially for GA-PLS models, RMSEC and RMSEP were <0.05%, R and R were >0.94, and RPD were both >4, indicating that both the models had good robustness and excellent prediction accuracy.
The present calibration models can be utilized to simultaneously determine the contents of liquirtin and glycyrrhizic acid in liquorice samples, and thus are of great help for rapid quality evaluation and control of liquorice.
建立同时测定甘草酸和甘草苷含量的校正模型，并考察变量选择方法。
采用高效液相色谱法（HPLC）测定甘草酸和甘草苷的含量作为参考值，利用近红外光谱（NIR）分析技术，将样品光谱与参考值相关联。采用偏最小二乘法（PLS）回归算法建立校正模型，并通过独立数据集测试，计算校正和预测的决定系数（R、R）、校正和预测的均方根误差（RMSEC、RMSEP）、校正和预测的平均绝对误差（MAEC、MAEP）和残差预测偏差（RPD）来评价模型。考察了变量重要性投影（VIP）、竞争自适应重加权采样（CARS）、蒙特卡罗无信息变量消除（MCUVE）、粒子群优化（PSO）和遗传算法（GA）等 5 种变量选择方法。
与原始全谱相比，甘草酸和甘草苷的定量模型均有明显改善，GA>PSO>CARS>MCUVE≅VIP>Full。特别是 GA-PLS 模型，RMSEC 和 RMSEP 均<0.05%，R 和 R 均>0.94，RPD 均>4，表明模型具有良好的稳健性和优异的预测精度。
本校正模型可用于同时测定甘草样品中甘草酸和甘草苷的含量，对甘草的快速质量评价和控制具有重要意义。
