DETERMINATION OF STARCH ADULTERATION IN GANODERMA LUCIDUM POLYSACCHARIDE BY NEAR INFRARED REFLECTANCE SPECTROSCOPY WITH PARTIAL LEAST SQUARES ALGORITHM

This paper reports the feasibility of determining the adulterated starch levels in Ganoderma lucidum polysaccharide using near infrared reflectance (NIR) spectroscopy coupled with multivariate methods. Multivariate calibration modeling procedures such as partial least squares (PLS), interval partial least squares, backward interval partial least squares, and forward interval partial least squares were applied to select the effective spectral regions that provided the lowest prediction error. According to the sum of squared differences and the root mean square error of prediction by the external validation, the optimal combinations of 8 spectral intervals were selected by forward interval partial least squares model. The optimal model was achieved with determination coefficient of 0.9962 and root mean square error of cross validation of 0.0252 in calibration set, and determination coefficient of 0.9957 and root mean square error of prediction of 0.0258 in validation set. The results indicated that NIR spectroscopy with forward interval partial least squares algorithm could be applied to determine the adulterated levels of starches in Ganoderma lucidum polysaccharide.