二阶导数光谱预处理在用FTIR/ATR方法定量测定 葡萄糖-6-磷酸和果糖-6-磷酸中的应用

利用傅里叶变换红外光谱(FTIR)和衰减全反射(ATR)技术,建立了一种葡萄糖-6-磷酸(G6P)和果糖-6-磷酸(F6P)同时定量测量的方法。作为磷酸己糖异构酶(PGI)反应系的模拟,G6P-F6P-Tris混合溶液被作为定量对象,并通过采集组分浓度已知的G6P-F6P-Tris混合溶液的光谱来建立定标方法。分别用吸光度光谱或二阶导数光谱,并分别用G6P,F6P,Tris的主要吸收峰(1086 cm^－1、1082 cm^－1、1065 cm^－1、1036 cm^－1和978 cm^－1)的数据或指纹领域(1 200~900cm^－1)全段数据,共进行了4种模式的多元线性回归分析,比较各种模式下浓度的预测值与实测值的相关系数、平均相对误差和交叉检验均方误差(RMSECV值)。结果表明,采用二阶导数光谱比采用吸光度光谱所得到的预测值具有更高的精度,而采用指纹领域全段数据比采用G6P,F6P,Tris主要吸收峰的数据所得到预测值具有更高的精度。

Application of second derivative spectrum prepares in quantification measuring glucose-6-phosphate and fructose-6-phosphate using a FTIR/ATR method

A simultaneous quantification measurement method of glucose-6-phosphate(G6P) and fructose-6-phosphate(F6P) was developed using a Fourier transform infrared(FTIR) spectrometer and attenuated total reflection(ATR) techniques.For applying the quantification method to the enzyme reaction system from G6P to F6P with phosphoglucose isomerase(PGI),the G6P-F6P-Tris mixture solution was used as the quantification object,and the spectra of the G6P-F6P-Tris mixture solutions which the concentrations were known were used as calibration spectra.Corresponding to the absorbance spectra or the second derivative spectra,selecting the value in main absorption peaks of 1 086 cm~(-1),(1 082 cm~(-1)),1 065 cm~(-1),1 036 cm~(-1) and 978 cm~(-1) or the data in the fingerprint region of 1 200~900 cm~(-1),four types of the multiple linear regressions analyses were performed.By comparing the related coefficients,the average relative error and the root mean square error cross validation(RMSECV) values for the calculated values and the actual values for the concentrations corresponding to all types,the enough accuracy could be achieved for the calculated values based on the data in the fingerprint region for the second derivative spectra.The calculated values corresponding to the second derivative spectra have the higher accuracy than ones corresponding to the absorbance spectra,and the calculated values based on the data in the fingerprint region have the higher accuracy than ones based on the data in main absorption peaks.