Table 1 Datasets for benchmarking the drug repurposing methods

From: Benchmarking heterogeneous network-based methods for drug repurposing

 

Dataset

Drugs

Diseases

Associations

Size

Sparsity

Ref

Public datasets

HDVD

219

34

455

7446

0.9389

59

LAGCN

269

598

18416

160862

0.8855

41

Fdataset

593

313

1933

185609

0.9895

60

Cdataset

663

409

2352

271167

0.9913

26

LRSSL

763

681

3051

519603

0.9941

61

Ydataset

1478

655

8448

968090

0.9912

5

New datasets

OMat-MechDB

89

150

271

13350

0.9797

This study

HSDN-MechDB

1279

616

3710

787864

0.9952

This study

  1. All drugs and diseases involve in at least one known drug-disease association. Column Size presents the number of data points in the association matrix, calculated by the multiplication of the number of drugs (column Drugs) and the number of diseases (column Diseases). Columns “Associations” presents the number of true associations. Sparsity is defined as the proportion of unknown associations, calculated by 1 - Associations/Size.
Back to article page