Skip to main content
Figure 3 | BMC Bioinformatics

Figure 3

From: Metabolite signal identification in accurate mass metabolomics data with MZedDB, an interactive m/z annotation tool utilising predicted ionisation behaviour 'rules'

Figure 3

Correlation analysis and mathematical relationships of explanatory signals discriminating healthy from diseased Brachypodium leaves. The left-hand panel displays the results of feature selection (all < P = 0.001, in descending rank order) in Random Forest classification models comparing FT-ICR-MS spectra of control Brachypodium distachyon leaves and plants 96 hours after challenge with a virulent strain of the rice blast fungus. The right hand panel shows example correlation clusters after a hierarchical cluster analysis (HCA) of the metabolome features (shown colour coded) in the left hand panel. The Pearson correlation coefficients are indicated for all combinations of ions in each cluster and the boxes below indicate accurate mass differences, predicted relationships and an annotation guide.

Back to article page