选用49份不同蛋白质和棉酚含量的陆地棉种质资源和188份陆地棉重组近交系为材料, 以多年份、 多地点种植收获的种子材料组成原始样品集, 分别对棉仁粉中蛋白质含量和棉酚含量进行化学测定, 采用改进的偏最小二乘法(Modified PLS)和(2, 4, 4, 1)的数学转换方法建立近红外反射光谱(NIRS)定标模型, 以寻找棉籽蛋白质含量和棉酚含量的快速测定方法。 结果表明, 蛋白质含量的定标决定系数(RSQ=0.933)和交叉检验决定系数(1-VR=0.929)较高, 定标标准误差(SEC=0.623)和交互校验标准误差(SECV=0.638)较小, 预测模型的建模效果较好, 可替代化学分析。 棉酚含量预测模型的RSQ, 1-VR, SEC和SECV分别为0.836, 0.811, 0.074和0.079, 模型预测效果略差于蛋白质模型, 但仍可用于棉仁粉中棉酚含量的测定。with Near Infrared Reflectance Spectroscopy

Near-infrared reflectance spectroscopy (NIRS) was used as a rapid andnondestructive method to determine the protein content and gossypol content incotton kernel powder samples, using 49 upland cotton (Gossypium hirsutum L.)germplasms and 188 recombinant inbred lines (RILs). The cottonseed samplesharvested from the upland cotton germplasms and RILs grown in different cottongrowing regions in different years were analyzed chemically for protein andgossypol contents, as well as scanned in the reflectance mode of a scanningmonochromator. Using ISI software for scanning and data analysis, protein andgossypol calibration equations were obtained with a standard normal variate +detrending scatter correction and a 2, 4, 4, 1 math treatment and modifiedpartial least square (MPLS) as the regression method. The protein contentcalibration results revealed that the multiple correlation coefficients (RSQ) andstatistic 1-variance ratio (1-VR) for the determination of protein content incottonseed kernels were 0.933 and 0.929, respectively, and its standard error ofcalibration (SEC) and standard error of cross validation (SECV) were 0.623 and0.638, respectively. As the calibration equations were judged by the calibrationRSQ (or 1-VR) and SEC (or SECV), the results indicated that NIRS is comparable tochemical methods in both accuracy and prediction and is reliable in thedetermination of protein content in cottonseed kernels. However, the RSQ, SEC, 1-VR and SECV for gossypol content determination of NIRS were 0.836, 0.811, 0.074and 0.079, respectively. Although it was weaker than that of protein content, theNIRS method is still good enough for the determination and prediction of thegossypol content in cottonseed kernels. Therefore, NIRS models were successfullydeveloped for protein content and gossypol content analysis of cotton kernelpowder sample in the present study and they could be introduced into the cottongermplasm evaluation and breeding program for the cottonseed quality improvement.Gossypol content; Protein content)科研专项经费项目(nyhyzx07-052)资助