Table 3 Comparison of model performance for the prediction of emotional disorders in girls at age 23

From: Genetic risk-dependent brain markers of resilience to childhood Trauma

PRSMDD cutoff

Girls with the higher PRSMDD (IMAGEN: age 19→age 23)

N

cases

Model performance (AUC)

GRBMR vs Network

Network vs Baseline

GRBMR

Network

Baseline

Median

128

63

0.749 ± 0.014

0.742 ± 0.013

0.737 ± 0.014

t = 4.731

t = 3.385

p = 3.35 × 10−5

p = 0.001

top 45%

114

54

0.759 ± 0.021

0.752 ± 0.022

0.742 ± 0.022

t = 2.762

t = 4.017

p = 0.006

p = 3.97 × 10−5

top 40%

100

46

0.755 ± 0.025

0.749 ± 0.026

0.740 ± 0.025

t = 2.841

t = 3.407

p = 0.005

p = 0.001

top 35%

88

42

0.746 ± 0.034

0.730 ± 0.032

0.721 ± 0.031

t = 3.756

t = 2.127

p = 0.001

p = 0.036

top 30%

74

38

0.756 ± 0.032

0.748 ± 0.038

0.733 ± 0.034

t = 3.078

t = 2.845

p = 0.003

p = 0.006

top 25%

62

34

0.758 ± 0.039

0.752 ± 0.042

0.741 ± 0.042

t = 2.487

t = 2.393

p = 0.016

p = 0.019

top 20%

50

28

0.762 ± 0.049

0.754 ± 0.052

0.749 ± 0.049

t = 2.792

t = 2.525

p = 0.007

p = 0.015

Mean ± sd

34

18

0.741 ± 0.045

0.732 ± 0.047

0.725 ± 0.051

t = 2.143

t = 2.301

p = 0.039

p = 0.028

  1. PRSMDD, polygenic risk score for major depressive disorder. GRBMR, genetic risk-dependent brain marker of resilience. AUC, area under the curve. The mean and the standard deviation established by repeating a 5-fold cross validation 10 times were reported before and after the ‘±’, respectively. Paired t tests were used to assess the significance of differences in AUC between models, with both t-values and p-values (two-sided) reported.