Skip to main content
Figure 4 | BMC Bioinformatics

Figure 4

From: A genetic algorithm-Bayesian network approach for the analysis of metabolomics and spectroscopic data: application to the rapid identification of Bacillus spores and classification of Bacillus species

Figure 4

Species all bacilli. (a) principal component analysis discriminating among species (all bacilli). v ¯ = = vegetative, s ¯ = = sporulated, 1 ¯ = = amyloliquefaciens, 2 ¯ = = cereus, 3 ¯ = = licheniformis, 4 ¯ = = megaterium, 5 ¯ = = sphaericus, 6 ¯ = = subtilis, 7 ¯ = = laterosporus. The bar over the labels indicates that each point represents an average over all similar cases on the data set. 1 v ¯ , … , 7 v ¯ and 1 s ¯ , … , 7 s ¯ represent group centers for PC scores and the numbers (which represent m/z intensities) are PC loadings. (b) Bayesian network model. Solid lines indicate a positive correlation between nodes, dashed lines indicate negative correlation and the thickness of the line indicates the strength of the correlation. Therefore the thicker the line is the stronger the correlation. The number beside the line shows the partial correlation coefficient for that correlation.

Back to article page