摘要
目的建立一种基于脂肪酸指纹的海带产地溯源的方法.方法通过对不同产地的海带中脂肪酸组成及含量进行测定,运用主成分分析方法从多种脂肪酸中筛选出8个特征脂肪酸优化判别变量,分别运用16种脂肪酸和8种优化脂肪酸为判别变量建立海带产地溯源模型.结果验证结果显示基于16种脂肪酸构建的判别模型原始样本成功判别率和交叉验证成功判别率分别为99.2%和95.8%,长岛、大连、荣成、霞浦的海带样品分别实现了100.0%、93.3%、90.0%、100.0%的交叉验证成功判别率,可以有效地满足对海带产地溯源的要求.基于8种优化脂肪酸构建的优化判别模型其变量数比原模型减少了50%,其原始样本成功判别率和交叉验证成功判别率分别为95.8%和95.0%,长岛、大连、荣成、霞浦的海带样品分别实现了90.0%、100.0%、93.3%、96.7%的交叉验证成功判别率,在减少了数据分析工作量的基础上依然能够有效满足海带产地溯源的要求.结论本研究通过对脂肪酸指纹进行分析,筛选出分组模型中的特征变量,建立海带的产地溯源模型,为今后探索海带产地溯源判定提供理论基础和技术指导.
Objective To establish a method for tracing the origin of Laminaria japonica based on fatty acid fingerprints.Methods By measuring the composition and content of fatty acids in Laminaria japonica from different origins,the principal component analysis method was used to select 8 characteristic fatty acid optimization discriminant variables from a variety of fatty acids,and 16 kinds of fatty acids and 8 kinds of optimized fatty acids were used as discriminant variables to establish a Laminaria japonica origin traceability model.Results The results showed that the successful discrimination rate of the original sample and the successful discrimination rate of cross validation of the discriminant model based on 16 kinds of fatty acids were 99.2%and 95.8%,respectively,the Laminaria japonica samples from Changdao,Dalian,Rongcheng,and Xiapu achieved 100.0%,93.3%,90.0%,and 100.0%successful cross validation discrimination rates,respectively,which could effectively meet the requirements for tracing the origin of Laminaria japonica.The number of variables in the optimized discriminant model constructed based on 8 kinds of optimized fatty acids was reduced by 50%compared to the original model,but the successful discriminant rate of the original sample and the successful discriminant rate of cross validation were 95.8%and 95.0%,respectively.The Laminaria japonica samples from Changdao,Dalian,Rongcheng,and Xiapu achieved 90.0%,100.0%,93.3%and 96.7%successful cross validation discrimination rates respectively,it could still effectively met the requirements for traceability of Laminaria japonica origin on the basis of reducing the workload of data analysis.Conclusion This study screens the characteristic variables in the grouping model by analyzing the fatty acid fingerprints,and establishs a source traceability model for Laminaria japonica,providing theoretical basis and technical guidance for future exploration of source traceability determination for Laminaria japonica.
作者
张华威
李英杰
曹伟
李佳蔚
周晓杰
邱子铭
陈婷婷
徐英江
孙利芹
崔艳梅
ZHANG Hua-Wei;LI Ying-Jie;CAO Wei;LI Jia-Wei;ZHOU Xiao-Jie;QIU Zi-Ming;CHEN Ting-Ting;XU Ying-Jiang;SUN Li-Qin;CUI Yan-Mei(Qixia Inspection and Testing Center,Yantai 265300,China;Yantai University College of Life Sciences,Yantai 264005,China;Shandong Marine Resource and Environment Research Institute,Yantai 264006,China)
出处
《实验室检测》
2023年第3期11-20,共10页
Laboratory Testing
基金
山东省藻类产业技术体系项目(SDAIT-26-05)
烟台市科技创新发展计划基础研究类项目(2022JCYJ035)
关键词
海带
脂肪酸指纹
产地溯源
判别分析
Laminaria japonica
fatty acid fingerprints
origin traceability
discriminant analysis
作者简介
张华威,硕士,高级工程师,主要研究方向为食品质量与安全。E-mail:vipzhw@126.com;通信作者:崔艳梅,硕士,高级工程师,主要研究方向为水产品质量安全。E-mail:cuiyanmei022l@163.com
