Skip to main content

Table 2 The prediction results of single-type DDIs

From: Predict multi-type drug–drug interactions in cold start scenario

Methods

Single-type DDIs

S1

S2

AUC

AUPR

F1

AUC

AUPR

F1

CSMDDI-SVD

0.8137

0.6121

0.3735

0.7266

0.4783

0.3698

CSMDDI-GAE

0.7223

0.4435

0.3735

0.6387

0.3390

0.3726

CSMDDI-RESCAL

0.8861

0.7270

0.4065

0.7709

0.5270

0.3751

TransE + RandomForest

0.6451

0.4371

0.3730

0.4391

0.1116

0.3679

DeepDDI(S)

0.7267

0.5087

0.3725

0.7254

0.4804

0.3745

DDIMDL(SP)

0.7556

0.5592

0.3732

0.7215

0.4729

0.3718

  1. (S) denotes that the inputs of the prediction method are chemical substructures
  2. (P) denotes that the inputs of the prediction method are drug-associated proteins
Back to article page

BMC Bioinformatics

ISSN: 1471-2105

Contact us