No abstract

We used near-infrared reflectance spectroscopy (NIRS) to assess biochemical parameters in whole oil flax seeds, regardless of differences in seed coat color of the samples. At the first stage of work, the set the task to develop calibration models for the MATRIX-I IR analyzer to determine the oil and moisture content in flax seeds. The carried out the research in the laboratory of biochemistry on brown and yellow seed samples of oil flax, grown in 2015-2020 in various agro-ecological conditions of the Russian Federation. We determined the oil content on an AMV 1006M NMR analyzer in accordance with the GOST 8.597- 2010 measurement procedure; we assessed the moisture content by the standard method of GOST 10856- 96. We used the results of determination of the oil and moisture content of the seeds of test lot in accordance with the accuracy indicator of the calibration of GOST 32749-2014 to verify the reliability of the developed models. We received the best indicators of the quality of calibration models (root-mean-square prediction error, coefficient of determination and the value of the residual deviation of prediction for the rank displayed on the graph) by determining the oil content (RMSEP = 0.27 %, R2 = 99.2 and RPD = 11.2) and moisture content (RMSEP = 0.06 %, R2 = 99.9 and RPD = 39). In the OPUS LAB program we developed the “Flax 51” method for mass analysis based on the developed calibration models for the determination of oil and moisture content in whole oil flax seeds (9-20 g) in a sample cell with a diameter of 51 mm. It enables the quick carrying out a preliminary assessment of the breeding material at a high speed – more than 120 samples in 7 hours without seed destruction.

МАСЛИЧНЫЕ КУЛЬТУРЫ. Научно-технический бюллетень Всероссийского научно-исследовательского института масличных культур. Вып. 4 (176), 2018 __________________________________________________________________________ __________________________________________________________________________

Spectroscopy of near infrared reflection (NIRS) was used for estimation of biochemical indicators in seeds of false flax. The purpose of our work was to develop calibrating models for IR-analyzer MATRIXI for determination of weight percentage of oil, linolenic and erucic acids contents in oil of seeds in unbroken seeds of false flax (winter and spring forms). The researches were conducted in the biochemistry laboratory on false flax samples cultivated in 2016–2020 in the different environments of the Russian Federation. Oil content was determined with NMR-analyzer АМV 1006М according to the technique described in the State Standard 8.597-2010, percentage contents of linolenic and erucic acids in oil was estimated on the gas chromatograph “Chromatech – Kristal 5000” with an automatic dipper on a capillary column SolGelWax 30 m × 0.25 mm × 0.5 µcm. The best indicators of quality of the calibrating models (root mean square error of prediction, coefficient of determination, and meaning of a residual deflection of prediction for a rank reflected on a figure) were obtained by oil content (RMSEP = 0.20%, R 2 = 99.3, and RPD = 12.3), linolenic acid content (RMSEP = 0.35%, R 2 = 98.8, and RPD = 9.2) and erucic acid content (RMSEP = 0.14%, R 2 = 85.7, and RPD = 2.6). In a program OPUS LAB, we received a method “False flax 51” based on the developed calibrating models for a routine analysis for determination of oil content, linolenic and erucic acids contents in oil in the unbroken seeds of false flax in an average (9–20 g) in a cuvette with diameter of 51 mm. this method allows conducting express-estimation of false flax seeds for breeding traits with performance of more than 100 sample per seven hours.

No abstract