Centro Nacional de Investigaciones de Café (Cenicafé), Caldas, Colombia.
J Agric Food Chem. 2009 Dec 9;57(23):11321-7. doi: 10.1021/jf902441n.
In a previous study, the effectiveness of chlorogenic acids, fatty acids (FA), and elements was compared for the discrimination of Arabica varieties and growing terroirs. Since FA provided the best results, the aim of the present work was to validate their discrimination ability using an extended experimental design, including twice the number of location x variety combinations and 2 years of study. It also aimed at understanding how the environment influences FA composition through correlation analysis using different climatic parameters. Percentages of correct classification of known samples remained very high, independent of the classification criterion. However, cross-validation tests across years indicated that prediction of unknown locations was less efficient than that of unknown genotypes. Environmental temperature during the development of coffee beans had a dramatic influence on their FA composition. Analysis of climate patterns over years enabled us to understand the efficient location discrimination within a single year but only moderate efficiency across years.
在之前的一项研究中，比较了绿原酸、脂肪酸（FA）和元素对于阿拉比卡品种和种植风土的区分效果。由于 FA 提供了最佳的结果，本研究的目的是使用扩展的实验设计来验证其区分能力，包括两倍的地点 x 品种组合和 2 年的研究。它还旨在通过使用不同的气候参数进行相关分析，了解环境如何影响 FA 组成。已知样本的正确分类百分比仍然非常高，与分类标准无关。然而，跨年度的交叉验证测试表明，对未知地点的预测效率低于对未知基因型的预测效率。咖啡豆发育过程中的环境温度对其 FA 组成有显著影响。对多年来气候模式的分析使我们能够理解在单一年份内的有效地点区分，但在跨年度时效率适中。
