Table 1 Number of FDA-approved and clinically tested drugs recovered for both drug-target datasets (i.e., DrugBank (DB) and DrugCentral (DC)) across the four investigated cancers.

From: Using predictive machine learning models for drug response simulation by calibrating patient-specific pathway signatures

Dataset

DB Prioritized

DB Approved (total)

DB Clinical trials (total)

DB Proportion of true positives (%)

DC Prioritized

DC Approved (total)

DC Clinical trials (total)

DC Proportion of true positives (%)

BRCA

129

8 (26)

23 (182)

31/129 (24.03%)

19

2 (14)

4 (115)

6/19 (31.57%)

LIHC

74

2 (5)

11 (50)

13/74 (17.56%)

19

1 (1)

2 (35)

3/19 (15.78%)

PRAD

68

2 (13)

18 (134)

20/68 (29.41%)

19

1 (7)

3 (84)

4/19 (21.05%)

KIRC

88

2 (8)

10 (44)

12/88 (13.63%)

26

3 (3)

2 (25)

5/26 (19.2%)

  1. In the first column for each drug-target dataset (“Prioritized”), we report the number of drugs that changed the predictions for at least 80% of the patients for each cancer type. The second column (“Approved”) reports the number of FDA-approved drugs among these prioritized drugs as well as the total number of FDA-approved/clinically tested drugs present in each dataset between parentheses. Similarly, the third column (“Clinical trials”) reports the number of drugs tested in clinical trials among the prioritized drugs and the total number of FDA-approved/clinically tested drugs between parentheses. Finally, the last column (“Proportion of true positives”) reports the proportion of true positives (both FDA-approved and clinically tested drugs) among the prioritized drugs.
Back to article page