Skip to main content

Table 5 Uncovering potential new DDIs from the 1D-CNN model

From: On the robustness of generalization of drug–drug interaction models

Drug1

Drug2

Side effect (SE)

SE freq (%)

Probability

Proof

Random

One-unseen

Both-unseen

Bupropion

Benazepril

Heart attack

25.8

0.951

0.182

e-1

Didanosine

Stavudine

Acidosis

13.6

0.872

0.068

2-2

Bupropion

Fluoxetine

Panic attack

10.5

0.868

0.100

e-3

Bupropion

Orphenadrine

Muscle spasm

24.2

0.865

0.161

e-4

Tramadol

Zolpidem

Diaphragmatic hernia

1.4

0.854

0.029

0.041

e-5

Paroxetine

Fluticasone

Muscle spasm

24.2

0.871

0.355

e-6

Paroxetine

Fluticasone

Fatigue

38.2

0.863

0.450

e-7

Fluoxetine

Methadone

Pain

37.6

0.769

e-8

Carboplatin

Cisplatin

Blood sodium decreased

23.1

0.749

0.097

0.080

e-9

Chlorthalidone

Fluticasone

High blood pressure

30.5

0.797

0.439

e-10

  1. The table shows the drug–drug potential side effects supported with the existence probability predicted with the literature link proof