Font Size: a A A

Optimization Of Near Infrared Spectroscopy Nondestructive Analysis Technology For Phytic Acid Contents In Cottonseeds And Its Application In The Researches Of Their Genotypic And Environmental Effects

Posted on:2018-05-03Degree:MasterType:Thesis
Country:ChinaCandidate:R B ZhaoFull Text:PDF
GTID:2393330548981704Subject:Crop
Abstract/Summary:PDF Full Text Request
Cotton is an important economic crop in our country.Cottonseed,which is rich in protein(27.83?45.60%)and fat(28.24?44.05%),is the main by-product of cotton production.Cottonseed protein is an important vegetable protein source,and cottonseed oil is one of the most important edible oil.However,the presence of phytic acid,an anti-nutritional factor in cottonseed,affects the value of comprehensive utilization of cottonseed.The selection of low-phytate cotton varieties is of great significance to improve the comprehensive utilization of cotton by-products.In present work,a near-infrared calibration model for nondestructive phytic acid contents in cottonseeds was established,and the interaction effects of genotype,location,and year of phytic acid contents in cottonseeds were studied with this technique,which provided method and basis for breeding of low-phytate cotton and comprehensive utilization of cottonseed.The main results were as follows:(1)Establishment of NIR calibration model for phytic acid contents in cottonseedsA total of 456 cottonseeds of different varieties planted in 11 test sites of six provinces and cities in the Yangtze River basin in 2014 and 2015 were selected as the materials.The phytic acid contents of cottonseeds were determined by ion chromatography.The spectral datas of cottonseeds were collected by Fourier transform near infrared spectroscopy.Savitzky-Golay smoothing and the first derivative,etc.were used to pretreat the original spectra.Linear partial least squares(PLS),nonlinear least square support vector machine(LS-SVM),weighted least squares support vector machines(WLS-SVM),and robust weighted least squares support vector machines(RWLS-SVM)methods were employed to establish the near-infrared calibration models of nondestructive phytic acid contents in cottonseeds.The results showed that SG smoothing + SNV + the first derivative method was the best method for spectral preprocessing.The optimal near-infrared calibration model obtained in this work for phytic acid contents in cottonseeds was WLS-SVM model,which had the highest values of predictive determination coefficient(R2)and residual predictive deviation(RPD),the lowest values of root-mean square error of coss-validation(RMSECV)and root-mean square error of prediction(RMSEP),R2 =0.9768,RPD = 6.5902,RMSECV = 0.0851 and RMSEP = 0.0801,respectively.The model had a good performance of prediction and the advantages of rapid,convenient,non-destructive,environmental protection,and other characteristics,the R2>0.9 and RPD>3,which could completely replace the traditional method for determination of phytic acid content.(2)Interaction effects of genotype and environment of phytic acid contents in cottonseedsCottonseeds of 11 different cotton varieties(Zheda-1,Zheda-2,Zheda-3,Zheda-4,Zheda-5,Zheda-6,Zheda-7,Zheda-8,Zheda-9,Zheda-10,Zheda-11)planted in 10 test sites(Lixian of Hunan,Yueyang of Hunan,Wuling of Hunan,Jiangshan of Zhejiang,Jinhua of Zhejiang,Hefei of Anhui,Wuhu of Anhui,Yancheng of Jiangsu,Nanjing of Jiangsu,Jiujiang of Jiangxi)in the Yangtze River basin in 2014 and 2015 were selected as the materials.The genotype,location,year and their interaction effects of phytic acid contents in cottonseeds were studied by NIR nondestructive analysis method developed in this research.The results showed that the average of phytic acid contents of 11 cotton varieties was 1.31%,and the range of variation was 1.04?1.53%,including1.04%for Zheda-8 and 1.53%for Zheda-3.The contents of phytic acid between different locations ranged from 1.06%to 1.67%,Lixian of Hunan had the lowest content of phytic acid,followed by Nanjing of Jiangsu.The phytic acid contents in 2014 and 2015 were 1.26%and 1.38%,respectively.The results of variance analysis revealed that the genotype,location,year,genotype × location,genotype x year,location × year,and interaction of the three factors of phytic acid contents in cottonseeds all reached the extremely significant level.The results of effect analysis indicated that the interaction effect of location x year of phytic acid contents in cottonseeds had the greatest variation,followed by the effect of location and the effect of year.The effect of genotype of phytic acid contents in cottonseeds was small,which indicated that the genetic improvement of phytic acid contents in cottonseeds was difficult.
Keywords/Search Tags:cottonseeds, phytic acid, genotype, environment, interaction effect, near-infrared spectroscopy analytical technique, calibration model
PDF Full Text Request
Related items