Comparison between wet analysis and near infrared reflectance spectroscopic analysis of quality traits in Brassica spp
Loading...
Date
2011
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
PAU
Abstract
The improvement in oil content, fatty acid profile and decreasing antinutritional factors are prerequisite for quality improvement in rapeseed-mustard. Estimation of quality parameters by reference methods is time consuming, expensive, labour intensive and sometimes hazardous. Near-infrared reflectance spectroscopy (NIRS) is an alternative technology that is non-destructive, accurate and inexpensive in long term use. The calibration of the instrument is mandatory for all the laboritories that install it. This requires accurate wet analysis of various biochemical quality traits, collection of reflectance spectra of samples, score generation, cross validation, prediction validation and development of calibration equation. The present study aims at developing calibration equation for the estimation of oil content, fatty acid profile, glucosinolates, proteins, moisture content in B. juncea and B. napus. Reflectance spectra of intact seeds of same samples were recorded on NIR spectrophotometer in the wavelength range of 400-2500nm. The NIRS manipulation for scanning, mathematical processing, and statistical analysis was performed with the WinISI software. Calibration equation was developed after cross validation using modified partial least-squares method. Internal validation was done to test the goodness of fit of the developed equation. The calibration equations for oil content (r2 0.83), oleic acid (r2 0.91), linoleic acid (r2 0.83), erucic acid (r2 0.88) and glucosinolates (r2 0.75) showed very good r2 in B. juncea and provided reliable estimations of these traits in the internal validation with r2 0.85, 0.77, 0.88, 0.94, 0.78 respectively. External validation also gives r2 0.72- 0.92 for oil, oleic acid, linoleic acid, erucic acid and glucosinolates The calibration equation for linolenic acid (r2 0.53) and moisture content (r20.61) was less reliable. In B. napus, oleic acid (r2 0.87), erucic acid (r2 0.87) gives good calibration and internal cross validation indicated equation valid for oleic acid (r2 0.91), linoleic acid (r2 0.71), erucic acid (r2 0.94). External cross validation results were similar to internal validation. The results indicated that NIRS could be used to rapidly determine oil content, oleic acid, linoleic acid, erucic acid and glucosinolates in intact B. juncea seeds and oleic acid, linoleic acid and erucic acid in B. napus.
Description
Keywords
null