Skip to main content

Table 4 Results of the DeepTGIN model and other compared models on the PDBbind2013 test set

From: DeepTGIN: a novel hybrid multimodal approach using transformers and graph isomorphism networks for protein-ligand binding affinity prediction

Models

R(\(\uparrow\))

RMSE(\(\downarrow\))

MAE(\(\downarrow\))

SD(\(\downarrow\))

CI(\(\uparrow\))

GraphDTA

0.674

1.661

1.287

1.660

0.740

DeepGLSTM

0.676

1.654

1.276

1.651

0.742

DeepDTAF

0.728

1.581

1.277

1.547

0.769

TEFDTA

0.736

1.536

1.210

1.522

0.762

IGN

0.782

1.411

1.135

1.406

0.788

GIGN

0.780

1.407

1.133

1.409

0.780

CAPLA

0.744

1.524

1.233

1.502

0.767

DeepTGIN

0.787

1.388

1.123

1.386

0.792

  1. \(\uparrow\) indicates that larger values indicate better performance, while \(\downarrow\) indicates that smaller values indicate better performance. The best results are shown in bold
Back to article page

Journal of Cheminformatics

ISSN: 1758-2946

Contact us