基于近红外光谱技术检测红曲米中的红曲色素

Determination of Monascus Pigments in Red Yeast Rice Using Near Infrared Spectroscopy

摘要:

采用近红外光谱技术结合化学计量学方法构建红曲米中红曲橙色素、红曲红色素、红曲黄色素的预测模型。分别采用多元线性回归（SMLR）、偏最小二乘回归（PLS）、主成分回归（PCR）构建所有色素组分的数学模型，以相关系数（R）、校正均方根误差（RMSEC）、预测均方根误差（RMSEP）、预测相对分析偏差（RPD）值来评价模型的综合性能。结果显示，MSC、SNV方法能够消除红曲米粉颗粒不均对光谱的散射影响；导数处理消除了基线漂移；对于红曲橙色素、红曲黄色素、红曲红色素三种模型均具有良好的稳定性；利用三种模型对未知红曲样品预测时，预测结果具有较高的线性，预测性能较好（RPD=2.86~5.39），可用于准确定量预测。结果表明近红外光谱技术可用于红曲色素的快速无损测定，为红曲米质量的智能化控制提供了新的途径。

Abstract:

NIR combined with chemometrics was proposed to predict orange, red and yellow pigments content in red yeast rice. Stepwise multiple linear regression (SMLR), partial least squares (PLS) and principal component regression (PCR) were used to built prediction models. Correlation coefficient of calibration (R), root mean square error of calibration (RMSEC), root mean square error of predication (RMSEP) and ratio of prediction to deviation (RPD) were suggested to evaluate the performance of models. The results showed that MSC and SNV could eliminate spectral scattering causing by uneven red yeast rice particles. Derivative treatment could eliminate the baseline drift. Three models for orange, red and yellow pigments all had good robustness. The three models were used to predict unknown monascus pigments, which all had better performance of prediction (RPD, 2.86~5.39). Therefore, the models could be used to accurately predict monascus pigments. The study shows that near-infrared spectroscopy technology has the potential beneficial for measuring the pigments content in red yeast rice online and conducive to intelligent quality control.