Models of near infrared spectroscopy of fatty acid contents in rapeseed

摘要

Content of oleic acid varied greatly in rapeseed. When rapeseeds with broad range of oleic acid content were used for calibration, models of near-infrared reflectance spectroscopy (NIRS) showed qualified accuracy. To further increase the accuracy of models, the present study narrowed the range of oleic content in calibration samples and only focused rapeseeds with oleic acid content ranging from 55 to 81%. Finally, external validation showed higher accuracy of the model for oleic acid than previous models, with the RSQ of 0.995. Meanwhile, models for linoleic and linolenic acids were also constructed, both showing high RSQ (0.929 and 0.907). These NIRS models could be used for fast determination of fatty acid contents in rapeseeds during agricultural and breeding practices.Practical applicationsThe NIRS method has advantages of rapidity, high efficiency, reliable results, no drug contamination, and no damage to materials. The models constructed in the present study could be used to determine contents of oleic, linoleic, and linolenic acids in rapeseeds. This method can determine several quality parameters with only a small amount of sample at the same time. It is of great significance to save manpower, materials and improve the analysis efficiency in plant breeding applications and rapeseed quality analysis. It is an effective detection method for screening rapeseed breeding materials, analyzing fatty acid components, and quality of rapeseeds.