Skip to main content

Table 2 10 fold cross validated performance of the base models in individual prediction stage on base valid set using their respective base features

From: CardioTox net: a robust predictor for hERG channel blockade based on deep learning meta-feature ensembles

Base features

MCC

NPV

ACC

PPV

SPE

SEN

AUC

DESC

0.689

0.813

0.845

0.870

0.868

0.822

0.911

MGF

0.620

0.805

0.810

0.817

0.794

0.826

0.888

MFP

0.683

0.830

0.841

0.855

0.837

0.845

0.915

FPeV

0.638

0.814

0.818

0.826

0.802

0.835

0.899

SeV

0.636

0.811

0.817

0.827

0.809

0.826

0.889

  1. Standard deviation value for each split for the above table is given in Additional file 1: S3
  2. Highest values are underlined
Back to article page

Journal of Cheminformatics

ISSN: 1758-2946

Contact us