Skip to main content
. 2018 Sep 5;8:13297. doi: 10.1038/s41598-018-31543-x

Figure 3.

Figure 3

Volatile metabolic fingerprints discriminate between CP and non-CP K. pneumoniae. (A) ROC curves for the discrimination between CP and non-CP K. pneumoniae generated using class probabilities for validation set samples. Green: RF; Orange: linear SVM; Blue: PLS-DA. Values in parentheses indicate AUROC. (B) Bee swarm plot depicting class probabilities (i.e., CP versus non-CP) for K. pneumoniae isolates, divided by ST. Red: CP K. pneumoniae; blue: non-CP K. pneumoniae. Points towards the top of the plot indicate a higher probability of classifying as CP, while those towards the bottom of the plot indicate a higher probability of classifying as non-CP. Horizontal bars represent the median class probability for each ST. ***: Previously unreported ST.