Classification of cultivated mussels from Galicia (Northwest Spain) with European Protected Designation of Origin using trace element fingerprint and chemometric analysis
被引:108
作者:
Costas-Rodriguez, M.
论文数: 0引用数: 0
h-index: 0
机构:
Univ Vigo, Fac Quim, Area Quim Analit, Dept Quim Analit & Alimentaria, Vigo 36310, SpainUniv Vigo, Fac Quim, Area Quim Analit, Dept Quim Analit & Alimentaria, Vigo 36310, Spain
Costas-Rodriguez, M.
[1
]
论文数: 引用数:
h-index:
机构:
Lavilla, I.
[1
]
论文数: 引用数:
h-index:
机构:
Bendicho, C.
[1
]
机构:
[1] Univ Vigo, Fac Quim, Area Quim Analit, Dept Quim Analit & Alimentaria, Vigo 36310, Spain
Galician mussel classification;
Linear discriminant analysis;
Soft independent modelling of class analogy;
Artificial neural networks;
Trace elements;
ICP-MS;
MYTILUS-GALLOPROVINCIALIS LMK;
NW IBERIAN PENINSULA;
GEOGRAPHICAL ORIGIN;
NEURAL-NETWORKS;
DNA MARKERS;
DISCRIMINATION;
SPECTROMETRY;
SEDIMENTS;
RIA;
D O I:
10.1016/j.aca.2010.03.003
中图分类号:
O65 [分析化学];
学科分类号:
070302 [分析化学];
摘要:
Inductively coupled plasma-mass spectrometry (ICP-MS) in combination with different supervised chemometric approaches has been used to classify cultivated mussels in Galicia (Northwest of Spain) under the European Protected Designation of Origin (PDO). 158 mussel samples, collected in the five rias on the basis of the production, along with minor and trace elements, including high field strength elements (HFSEs) and rare earth elements (REEs), were used with this aim. The classification of samples was achieved according to their origin: Galician vs. other regions (from Tarragona, Spain, and Ethang de Thau, France) and between the Galician Rias. The ability of linear discriminant analysis (LDA), soft independent modelling of class analogy (SIMCA) and artificial neural network (ANN) to classify the samples was investigated. Correct assignations for Galician and non-Galician samples were obtained when LDA and SIMCA were used. ANNs were more effective when a classification according to the ria of origin was to be applied. (C) 2010 Elsevier B.V. All rights reserved.