Skip to main content

Table 1 Performance evaluation of methods in identifying misclassified samples

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

Misclassification

Median AUC ROCa

Median AUC PRb

PheLEx

Rekaya

PheLEx

Rekaya

1%

0.795

0.527

0.276

0.0323

3%

0.783

0.523

0.418

0.09

5%

0.773

0.522

0.471

0.144

8%

0.776

0.514

0.565

0.205

10%

0.768

0.514

0.645

0.240

  1. aAUC ROC = Area under Receiver Operating Characteristic curve
  2. bAUC PR = Area under Precision Recall curve