Please use this identifier to cite or link to this item: http://hdl.handle.net/1822/3560

TitlePortuguese regional differences in the wine grapes mycoflora
Author(s)Serra, Rita
Lourenço, Anália
Belo, Orlando
Venâncio, Armando
Issue date2005
PublisherUniversidade do Minho. Micoteca
CitationTRENDS IN MYCOLOGY, Braga, 2005 - “Trends in mycology : one day meeting… 23rd September 2005”. Braga : Micoteca da Universidade do Minho, [2005].
Abstract(s)The recent discovery of mycotoxins in wine, in particular ochratoxin A, caused concern and motivated an extensive survey to the mycoflora of Portuguese grapes. It is known that the mycoflora of agricultural commodities can vary according to the geographical origin, and therefore, regional differences in the mycoflora of Portuguese were investigated. Four regions were selected for a 3- year study: Alentejo, Douro, Ribatejo and Vinhos Verdes. The mycoflora of grapes was evaluated by plating methods. A total of 32 grape samples were taken, of 50 berries each. The differences in the mycoflora of grapes between regions were analyzed using the non-parametric test Kruskal-Wallis H. Ostensibly, the classification of the grapes into their geographical origin based on its mycoflora was attempted using a decision tree algorithm (C4.5) based on the Shannon Information Theory. Of the 27 fungal genera identified, 3 varied its incidence significantly according to the region of origin of the samples: Aspergillus, Botrytis and Ulocladium. The only species that varied significantly its frequency between regions was A. niger aggregate. Six Penicillium species differed significantly between regions: P. brevicompactum, P. citrinum, P. glabrum/spinulosum, P. expansum, P. implicatum and P. thomii. Using decision trees it was possible to classify successfully 91% of the samples according to 3 sample classes: Vinhos Verdes, Douro and South samples (Alentejo and Ribatejo). The classification was based on the incidence of A. niger and P. thomii in the grape samples. The estimated predictive ability of the model in the 3 classes was 82%. The data presented here indicate that grapes are consistently exposed to a particular mycoflora that varies according its geographical origin, which may be of importance to establishing risk areas for mycotoxin contamination of grapes and wine.
TypeAbstract
URIhttp://hdl.handle.net/1822/3560
Peer-Reviewedyes
AccessOpen access
Appears in Collections:CEB - Resumos em Livros de Atas / Abstracts in Proceedings

Files in This Item:
File Description SizeFormat 
Trends_in_Mycology.03[1].pdf12,74 kBAdobe PDFView/Open

Partilhe no FacebookPartilhe no TwitterPartilhe no DeliciousPartilhe no LinkedInPartilhe no DiggAdicionar ao Google BookmarksPartilhe no MySpacePartilhe no Orkut
Exporte no formato BibTex mendeley Exporte no formato Endnote Adicione ao seu ORCID