Skip to main content

Table 3 Prospective evaluation performance

From: Industry-scale application and evaluation of deep learning for drug target prediction

Algorithm

Metric

AstraZeneca

Janssen

DNN

ROC-AUC

0.70 ± 0.14

0.66 ± 0.16

Kappa

0.20 ± 0.19

0.15 ± 0.19

F1

0.42 ± 0.26

0.43 ± 0.24

XGB

ROC-AUC

0.67 ± 0.15

0.64 ± 0.15

Kappa

0.13 ± 0.17

0.10 ± 0.17

F1

0.35 ± 0.25

0.39 ± 0.27

MF

ROC-AUC

0.68 ± 0.15

0.64 ± 0.15

Kappa

0.12 ± 0.15

0.09 ± 0.14

F1

0.35 ± 0.29

0.38 ± 0.30

  1. Prospective evaluation performance values (mean and standard deviation across targets) for the considered machine learning algorithms