Skip to main content

Table 3 Prospective evaluation performance

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

AlgorithmMetricAstraZenecaJanssen
DNNROC-AUC0.70 ± 0.140.66 ± 0.16
Kappa0.20 ± 0.190.15 ± 0.19
F10.42 ± 0.260.43 ± 0.24
XGBROC-AUC0.67 ± 0.150.64 ± 0.15
Kappa0.13 ± 0.170.10 ± 0.17
F10.35 ± 0.250.39 ± 0.27
MFROC-AUC0.68 ± 0.150.64 ± 0.15
Kappa0.12 ± 0.150.09 ± 0.14
F10.35 ± 0.290.38 ± 0.30
  1. Prospective evaluation performance values (mean and standard deviation across targets) for the considered machine learning algorithms