January 10-14, 2004
Town & Country Convention Center
San Diego, CA
Workshop: Forest Trees
Microarrays were used to identify genes involved in dormancy and frost tolerance in forest tree seedlings. The resulting dataset included expression profiles of more than 1500 genes in autumn and winter in buds from Pinus sylvestris grown in Scotland or Denmark, and also physiological data such as REL, SEL and days to budbreak (DBB).
To be able to select genes involved in dormancy and cold hardiness development it was necessary to analyse the microarray results in comparison with the physiological data. A direct comparison would only reveal genes whose expression profiles followed the REL, SEL or DBB curves. However, relevant genes may have a different expression profile. To solve this we developed a pairwise comparative analysis of expressional and physiological data.
The robustness of the resulting selection of relevant genes was shown, by using two different statistical approaches. Using this method we selected genes whose expression profiles were correlated to development of cold hardiness in pine buds. By including samples from trees grown in different climates we were able to zoom in on the actual physiological change and distinguish that from random climate effects. The presented method is not only based on statistics but includes biological knowledge as well and is therefore highly suitable to identify relevant genes.
This work is part of the COLDTREE project, an EU funded programme that aims at unraveling the molecular processes underlying cold-hardiness and dormancy in forest tree species. In the presentation a short update of the COLDTREE results will be included.
W87ROBUST SELECTION OF RELEVANT GENES FROM MICROARRAY EXPERIMENTS BY INCLUDING PHYSIOLOGICAL DATA IN A PAIRED COMPARATIVE ANALYSIS
Eric Boer1
, Lonneke Van der Geest2
, Peter Brønnum3
, Michael Perks4
, Monique Van Wordragen1
Return to Previous Page or Intl-PAG Homepage