Towards developing reliable models of leaf area on grapevines (Vitis vinifera L.)

A field study in three vineyards in southern Queensland (Australia) was carried out to develop predictive models for individual leaf area and shoot leaf area of two cultivars (Cabernet Sauvignon and Shiraz) of grapevines (Vitis Vinifera L.). Digital image analysis was used to measure leaf vein length and leaf area. Stepwise regressions of untransformed and transformed models consisting of up to six predictor variables for leaf area and three predictor variables for shoot leaf area were carried out to obtain the most efficient models. High correlation coefficients were found for log10 transformed individual leaf and shoot leaf area models. The significance of predictor variables in the models varied across vineyards and cultivars, demonstrating the discontinuous and heterogeneous nature of vineyards. The application of this work in a grapevine modeling environment and in a dynamic vineyard management context are discussed. Sample sizes for quantification of individual leaf areas and areas of leaves on shoots are proposed based on target margins of errors of sampled data.

Modulation of higher-plant NAD(H)-dependent glutamate dehydrogenase activity in transgenic tobacco via alteration of beta subunit levels

Glutamate dehydrogenase (GDH; EC 1.4.1.2-1.4.1.4) catalyses in vitro the reversible amination of 2-oxoglutarate to glutamate. In vascular plants the in vivo direction(s) of the GDH reaction and hence the physiological role(s) of this enzyme remain obscure. A phylogenetic analysis identified two clearly separated groups of higher-plant GDH genes encoding either the alpha- or beta-subunit of the GDH holoenzyme. To help clarify the physiological role(s) of GDH, tobacco (Nicotiana tabacum L.) was transformed with either an antisense or sense copy of a beta-subunit gene, and transgenic plants recovered with between 0.5- and 34-times normal leaf GDH activity. This large modulation of GDH activity (shown to be via alteration of beta-subunit levels) had little effect on leaf ammonium or the leaf free amino acid pool, except that a large increase in GDH activity was associated with a significant decrease in leaf Asp (similar to 51%, P=0.0045). Similarly, plant growth and development were not affected, suggesting that a large modulation of GDH beta-subunit titre does not affect plant viability under the ideal growing conditions employed. Reduction of GDH activity and protein levels in an antisense line was associated with a large increase in transcripts of a beta-subunit gene, suggesting that the reduction in beta-subunit levels might have been due to translational inhibition. In another experiment designed to detect post-translational up-regulation of GDH activity, GDH over-expressing plants were subjected to prolonged dark-stress. GDH activity increased, but this was found to be due more likely to resistance of the GDH protein to stress-induced proteolysis, rather than to post-translational up-regulation.

Cluster analysis of high-dimensional data: A case study

Normal mixture models are often used to cluster continuous data. However, conventional approaches for fitting these models will have problems in producing nonsingular estimates of the component-covariance matrices when the dimension of the observations is large relative to the number of observations. In this case, methods such as principal components analysis (PCA) and the mixture of factor analyzers model can be adopted to avoid these estimation problems. We examine these approaches applied to the Cabernet wine data set of Ashenfelter (1999), considering the clustering of both the wines and the judges, and comparing our results with another analysis. The mixture of factor analyzers model proves particularly effective in clustering the wines, accurately classifying many of the wines by location.