Table 3 Performance of the logistic regression (LR) classifier on the aggregate and site-level analyses.

From: Prediction of lithium response using genomic data

Centre (LR+/LR−; %)

AUC

Accuracy

F1

NPV

PPV

Sensitivity

Specificity

Kappa

ALL (29/71)

0.57 (0.55, 0.6)

0.7 (0.7, 0.71)

0.08 (0.05, 0.1)

0.72 (0.71, 0.72)

0.36 (0.27, 0.46)

0.04 (0.03, 0.06)

0.97 (0.97, 0.98)

0.02 (− 0.01, 0.04)

Halifax (45/55)

0.62 (0.58, 0.65)

0.61 (0.57, 0.64)

0.34 (0.24, 0.44)

0.59 (0.57, 0.62)

0.68 (0.58, 0.78)

0.23 (0.16, 0.31)

0.91 (0.87, 0.95)

0.15 (0.07, 0.24)a

Würzburg (17/83)

0.6 (0.51, 0.69)

0.85 (0.84, 0.86)

0.23 (0.12, 0.34)

0.85 (0.84, 0.86)

0.8 (0.41, 1.0)

0.13 (0.07, 0.2)

1.0 (1.0, 1.0)

0.2 (0.1, 0.3)a

Barcelona (27/73)

0.4 (0.26, 0.53)

0.73 (0.72, 0.74)

0.0 (0.0, 0.0)

0.73 (0.72, 0.74)

0.0 (0.0, 0.0)

0.0 (0.0, 0.0)

1.0 (1.0, 1.0)

0.0 (0.0, 0.0)

Cagliari (28/72)

0.49 (0.46, 0.52)

0.72 (0.72, 0.72)

0.0 (0.0, 0.0)

0.72 (0.72, 0.72)

0.0 (0.0, 0.0)

0.0 (0.0, 0.0)

1.0 (1.0, 1.0)

0.0 (0.0, 0.0)

Geneva (23/77)

0.49 (0.28, 0.7)

0.77 (0.74, 0.8)

0.0 (0.0, 0.0)

0.77 (0.74, 0.8)

0.0 (0.0, 0.0)

0.0 (0.0, 0.0)

1.0 (1.0, 1.0)

0.0 (0.0, 0.0)

Japan (24/76)

0.6 (0.45, 0.75)

0.76 (0.75, 0.77)

0.0 (0.0, 0.0)

0.76 (0.75, 0.77)

0.0 (0.0, 0.0)

0.0 (0.0, 0.0)

1.0 (1.0, 1.0)

0.0 (0.0, 0.0)

Mayo (23/77)

0.32 (0.21, 0.44)

0.77 (0.75, 0.78)

0.0 (0.0, 0.0)

0.77 (0.75, 0.78)

0.0 (0.0, 0.0)

0.0 (0.0, 0.0)

1.0 (1.0, 1.0)

0.0 (0.0, 0.0)

Paris (18/82)

0.4 (0.29, 0.51)

0.82 (0.81, 0.83)

0.0 (0.0, 0.0)

0.82 (0.81, 0.83)

0.0 (0.0, 0.0)

0.0 (0.0, 0.0)

1.0 (1.0, 1.0)

0.0 (0.0, 0.0)

Poznan (48/52)

0.62 (0.44, 0.8)

0.51 (0.39, 0.62)

0.28 (0.05, 0.51)

0.52 (0.44, 0.61)

0.36 (0.08, 0.64)

0.24 (0.03, 0.45)

0.76 (0.6, 0.92)

0.0 (− 0.23, 0.23)

Romania (21/79)

0.57 (0.51, 0.64)

0.79 (0.78, 0.8)

0.0 (0.0, 0.0)

0.79 (0.78, 0.8)

0.0 (0.0, 0.0)

0.0 (0.0, 0.0)

1.0 (1.0, 1.0)

0.0 (0.0, 0.0)

San Diego (11/89)

0.54 (0.5, 0.58)

0.89 (0.88, 0.9)

0.0 (0.0, 0.0)

0.89 (0.88, 0.9)

0.0 (0.0, 0.0)

0.0 (0.0, 0.0)

1.0 (1.0, 1.0)

0.0 (0.0, 0.0)

Sweden (25/55)

0.52 (0.46, 0.58)

0.55 (0.54, 0.56)

0.22 (0.11, 0.33)

0.56 (0.54, 0.57)

0.49 (0.41, 0.58)

0.15 (0.06, 0.25)

0.88 (0.81, 0.94)

0.03 (− 0.0, 0.07)

Sydney (20/80)

0.32 (0.18, 0.47)

0.78 (0.75, 0.81)

0.0 (0.0, 0.0)

0.79 (0.76, 0.82)

0.0 (0.0, 0.0)

0.0 (0.0, 0.0)

0.98 (0.94, 1.0)

− 0.03 (− 0.07, 0.02)

Taiwan (14/86)

0.49 (0.29, 0.69)

0.86 (0.84, 0.88)

0.0 (0.0, 0.0)

0.86 (0.84, 0.88)

0.0 (0.0, 0.0)

0.0 (0.0, 0.0)

1.0 (1.0, 1.0)

0.0 (0.0, 0.0)

  1. Table columns represent different classification statistics and values represent the mean of each statistic over five folds along with an empirical 95% confidence interval. In the column listing respective centres, we have included the percentage of lithium responders (LR+) and non-responders (LR−) in the format %LR+/%LR− to provide context for the classification results. Abbreviations: all sites (ALL; i.e. aggregate analysis), area under the receiver operating characteristic curve (AUC), positive predictive value (PPV), negative predictive value (NPV), F-1 score (F1).
  2. aKappa value was found to have a p value less than 0.01 in comparison to the null classifier.
Back to article page