Skip to main content
Fig. 2 | BMC Bioinformatics

Fig. 2

From: Identifying novel associations in GWAS by hierarchical Bayesian latent variable detection of differentially misclassified phenotypes

Fig. 2

Performance comparison between PheLEx and other misclassification extraction methods in identifying misclassified samples in simulations. Misclassification extraction methods: PheLEx (red), PheLEx-mm (purple), PheLEx-mh (teal), and Rekaya (blue). a Box plots showing area under Receiver Operating Characteristic (ROC) curves (AUC ROC) (y-axis) for identifying misclassified samples across simulations against increasing misclassification rates (x-axis) for misclassification extraction methods. b Box plots showing area under Precision-Recall (PR) curves (AUC PR) (y-axis) for identifying misclassified samples across simulations against increasing misclassification rates (x-axis) for misclassification extraction methods. c Box plots for running time (hours) (y-axis) is shown for misclassification extraction methods (x-axis) across simulations

Back to article page