Identification of genetic variation in Brassica napus seeds for tocopherol content and composition using near‐infrared spectroscopy technique

Tocopherol is an essential fat‐soluble nutrient for humans. Increasing the tocopherol content in Brassica napus seeds can add value to rapeseed vegetable oil; this has become an important breeding target. However, there is no efficient and non‐destructive method for selecting rapeseed accessions with high tocopherol contents. Here, we report the first near‐infrared reflectance spectroscopy (NIRS)‐based technique for predicting rapeseed tocopherol content. Individual seed tocopherol compositions were estimated from 373 rapeseed genotypes of different origins. This method and chemical methods produced comparable predicted values of the tocopherol constituents in the seeds. Three equations were generated for the prediction of tocopherol content by using a modified partial least squares (MPLS) model. The total tocopherol content for the determination coefficient of cross‐validation (R²_cv) (0.74), determination coefficient (RSQ) (0.76) and one minus the ratio of unexplained variance to total variance (1‐VR) (0.65) values indicates a strong correlation between the calibration and validation sets. Overall, our model confirmed the NIRS method as feasible for predicting tocopherol content in rapeseed and as an efficient screening tool for future breeding programs.