Nie Z, Han J, Liu T, Liu X
Department of Grassland Science, College of Animal Science and Technology, China Agricultural University, Beijing 10094, China.
J Dairy Sci. 2008 Jun;91(6):2361-9. doi: 10.3168/jds.2008-0985.
The object of this study was to explore the potential for support vector machine (SVM) to improve the precision of predicting protein fractions by near infrared reflectance spectroscopy (NIRS). Generally, most protein fractions determined in Cornell Net Carbohydrate and Protein System (CNCPS), especially the neutral detergent insoluble protein (NDFCP) and acid detergent insoluble protein (ADFCP), could not be accurately predicted by the commonly used partial least squares (PLS) method. A recently developed chemometric method, SVM, was applied in NIRS prediction of alfalfa protein fractions in this study. Two hundred thirty alfalfa samples were scanned on a near infrared reflectance spectrophotometer, and analyzed for crude protein (CP), true protein precipitated in tungstic acid (TCP), borate-phosphate buffer-insoluble protein (BICP), NDFCP, and ADFCP. These 5 laboratory proteins and the CNCPS protein fractions A, B1, B2, B3, and C were predicted by NIRS using the PLS and SVM methods. According to PLS-NIRS regression, CP, TCP, BICP, A, and B2 obtained the determination coefficient of prediction (R(p)(2)) of 0.96, 0.91, 0.94, 0.94, and 0.93, and the ratios of standard deviation of prediction samples: standard error of prediction samples (RPD) values were 5.07, 3.31, 3.98, 3.96, and 3.91. Neutral detergent insoluble protein, ADFCP (fraction C), B1, and B3 were predicted with R(p)(2) of 0.75, 0.83, 0.30, and 0.62, and RPD values of 1.98, 2.42, 1.20, and 1.62; Calibrated by the SVM-NIRS method, R(p)(2) values of CP, TCP, BICP, NDFCP, ADFCP(C), A, and B2 achieved 0.99, 0.97, 0.97, 0.90, 0.93, 0.97, and 0.97, respectively. The RPD values of those fractions were 8.68, 8.26, 6.11, 3.08, 3.69, 5.97, and 5.81, respectively. The R(p)(2) and RPD values of fractions B1 and B3 were 2.67 and 0.87 (B1) and 2.51 and 0.75 (B3) directly predicted by SVM-NIRS model. In this study, the chemical analysis results of B1 and B3 were also correlated with calculated results from TCP-BICP and NDFCP-ADFCP, which were predicted by SVM-NIRS models. The B1 protein fraction achieved R(p)(2) and RPD values of 0.87 and 3.61, whereas values for B2 were 0.75 and 2.00. Data suggested that use of SVM methods in NIRS technology could improve the accuracy of predicting protein fractions. This study showed the potential of increasing the NIRS prediction accuracy to a level of practical use for all protein fractions, except B3.
本研究的目的是探索支持向量机（SVM）提高近红外反射光谱法（NIRS）预测蛋白质组分精度的潜力。一般来说，康奈尔净碳水化合物和蛋白质体系（CNCPS）中测定的大多数蛋白质组分，尤其是中性洗涤不溶性蛋白质（NDFCP）和酸性洗涤不溶性蛋白质（ADFCP），无法通过常用的偏最小二乘法（PLS）准确预测。本研究将一种最近开发的化学计量学方法——支持向量机应用于苜蓿蛋白质组分的近红外光谱预测。在近红外反射分光光度计上扫描了230个苜蓿样品，并分析了粗蛋白（CP）、钨酸沉淀真蛋白（TCP）、硼酸盐 - 磷酸盐缓冲液不溶性蛋白（BICP）、NDFCP和ADFCP。使用PLS和SVM方法通过近红外光谱预测这5种实验室蛋白质以及CNCPS蛋白质组分A、B1、B2、B3和C。根据PLS - NIRS回归，CP、TCP、BICP、A和B2的预测决定系数（R(p)(2)）分别为0.96、0.91、0.94、0.94和0.93，预测样品标准差与预测样品标准误的比值（RPD）分别为5.07、3.31、3.98 和3.96。中性洗涤不溶性蛋白质、ADFCP（组分C）、B1和B3的预测R(p)(2)分别为0.75、0.83、0.30和0.62，RPD分别为1.98、2.42、1.20和1.62；通过SVM - NIRS方法校准后，CP、TCP、BICP、NDFCP、ADFCP(C)、A和B2的R(p)(2)值分别达到0.99、0.97、0.97、0.90、0.93、0.97和0.97。这些组分的RPD值分别为8.68、8.26、6.11、3.08、3.69、5.97和5.81。SVM - NIRS模型直接预测的B1和B3组分的R(p)(2)和RPD值分别为（B1）2.67和0.87以及（B3）2.51和0.75。在本研究中，B1和B3的化学分析结果也与通过SVM - NIRS模型预测的TCP - BICP和NDFCP - ADFCP的计算结果相关。B1蛋白质组分的R(p)(2)和RPD值分别为0.87和3.61，而B2的值分别为0.75和2.00。数据表明，在近红外光谱技术中使用支持向量机方法可以提高预测蛋白质组分的准确性。本研究表明，除B3外，将近红外光谱预测准确性提高到实际应用水平具有潜力。
