Skip to main content

Table 1 Summary of topic model for the 26 known ALF drugs in LiverTox database.

From: Mining hidden knowledge for drug safety assessment: topic modeling of LiverTox as a case study

Known ALF drugs 1st topic 2nd topic 3rd topic 4th topic
  Topic ID Prob. Topic ID Prob. Topic ID Prob. Topic ID Prob.
Acetaminophen 37 4.7E-01 29 3.6E-01 4 1.4E-01 6 1.6E-02
Amoxicillin/clavulanate 37 6.3E-01 4 1.9E-01 15 1.2E-01 20 6.1E-02
Atorvastatin 31 8.3E-01 39 1.7E-01 40 3.3E-05 38 3.3E-05
Benzbromarone* 28 4.4E-01 37 4.2E-01 11 1.4E-01 40 3.8E-05
Carbamazepine 3 4.6E-01 35 3.5E-01 38 1.9E-01 40 6.7E-05
Ciprofloxacin 15 7.7E-01 14 2.3E-01 40 2.8E-05 39 2.8E-05
Cyclophosphamide 31 6.4E-01 19 2.9E-01 40 6.2E-02 39 5.2E-05
Dapsone 7 8.0E-01 31 2.0E-01 40 3.3E-05 39 3.3E-05
Diclofenac 37 8.0E-01 20 1.4E-01 23 3.3E-02 22 2.2E-02
Disulfiram 37 3.6E-01 38 2.6E-01 28 2.2E-01 7 1.6E-01
Doxycycline 37 5.9E-01 17 3.3E-01 22 7.2E-02 40 3.6E-05
Ethambutol 28 3.9E-01 10 3.5E-01 7 1.7E-01 24 8.7E-02
Halothane 23 7.1E-01 37 2.8E-01 16 6.5E-03 40 1.6E-05
Ibuprofen 31 3.2E-01 21 2.9E-01 37 2.0E-01 16 1.3E-01
Isoniazid 37 6.2E-01 23 1.2E-01 10 9.5E-02 16 6.7E-02
Methyldopa 37 7.9E-01 20 2.1E-01 40 5.2E-05 39 5.2E-05
Naproxen 37 6.2E-01 30 1.8E-01 20 1.6E-01 2 3.6E-02
Nefazodone 37 4.2E-01 26 2.7E-01 24 1.8E-01 28 1.3E-01
Nitrofurantoin 37 1.0E+00 40 4.6E-05 39 4.6E-05 38 4.6E-05
Phenytoin 7 3.0E-01 37 2.5E-01 3 2.1E-01 35 1.6E-01
Simvastatin 31 4.9E-01 37 4.4E-01 34 6.1E-02 26 1.0E-02
Sulfamethoxazole/
trimethoprim
7 8.3E-01 28 1.7E-01 40 3.5E-05 39 3.5E-05
Sulfasalazine 7 6.8E-01 37 3.2E-01 40 3.2E-05 39 3.2E-05
Tolcapone* 37 6.4E-01 25 2.7E-01 3 8.7E-02 40 3.1E-05
Troglitazone* 37 4.6E-01 31 4.2E-01 1 1.1E-01 40 2.2E-05
Valproate 35 4.2E-01 38 3.1E-01 26 2.1E-01 14 5.2E-02
  1. *These three drugs are withdrawn from the market due to the drug-induced ALF.
  2. Topic-37 is the ALF-Topic.
  3. Topic ID: serial number of topics in topic model.
  4. Prob.: conditional probability of each topic to a drug.
  5. Topic model ranks topics in accordance to their probability to each drug.