Magdas Dana Alina, Dehelean Adriana, Feher Ioana, Radu Stelian
a National Institute for Research and Development of Isotopic and Molecular Technologies , Cluj-Napoca , Romania.
Isotopes Environ Health Stud. 2017 Dec;53(6):610-619. doi: 10.1080/10256016.2017.1335722. Epub 2017 Jun 8.
Two marker combinations were used for the differentiation of organically produced from conventionally produced potatoes and also for the geographical origin identification. Fifty-seven samples (from Romanian local producers or imported) were analysed from the stable isotopic (isotope ratio mass spectrometry) and elemental profile (inductively coupled plasma mass spectrometry) point of view. In order to assess the best marker combination, both isotopic and elemental experimental results were subject to chemometric analysis. The statistical tests performed were ANOVA test, Pearson correlation, principal component analysis and linear discriminant analysis (LDA). For a more comprehensive differentiation between organic vs. conventional potato samples, LDA was applied, and 94.7 % of original cases were correctly classified and the percentage obtained in cross-validation procedure was 91.2 %. Regarding the geographic origin classification, LDA provided an initial classification of 96.5 %, while for cross-validation the percentage was 87.7. LDA found δN, Cd, Ca, Cu and Zn as best discrimination markers between organically and conventionally grown potatoes. The strongest predictors for Romania vs. foreign geographical areas along LDA were seen to be Ca, P, Co, Ni and δC.
两种标记组合用于区分有机生产和传统生产的土豆，同时也用于地理来源鉴定。从稳定同位素（同位素比率质谱法）和元素谱（电感耦合等离子体质谱法）的角度分析了57个样品（来自罗马尼亚当地生产商或进口）。为了评估最佳标记组合，对同位素和元素实验结果都进行了化学计量分析。所进行的统计测试包括方差分析、皮尔逊相关性分析、主成分分析和线性判别分析（LDA）。为了更全面地区分有机土豆样品和传统土豆样品，应用了线性判别分析，原始案例中有94.7%被正确分类，交叉验证过程中获得的百分比为91.2%。关于地理来源分类，线性判别分析的初始分类率为96.5%，而交叉验证的百分比为87.7%。线性判别分析发现δN、Cd、Ca、Cu和Zn是区分有机种植土豆和传统种植土豆的最佳判别标记。沿着线性判别分析，罗马尼亚与外国地理区域之间最强的预测因子是Ca、P、Co、Ni和δC。
