Skip to main content

Table 1 Examples of some drug-target candidates generated by our predictor

From: Leveraging 3D chemical similarity, target and phenotypic data in the identification of drug-protein and drug-adverse effect associations

TCa Similar drug in ChEMBL (ATC category)b Drug candidate (ATC category) 3D D-Tc Target EF and q-valuesd
0.30 Diclofenac (antiinflammatory agent, non-steroid) Carbamazepine (carboxamide deriv., antiepileptic) 0.83 Gamma-secretase EF = 3.17 q < .05
0.20 Phenytoin (hydantoin deriv., antiepileptic) Venlafaxine (antidepressant) 0.82 Aquaporin-4 EF = 2.71 q < .05
0.65 Ondansetron (serotonin antagonist, antiemetic-antinauseant) Molindone (indole deriv., antipsychotic) 0.79 5-HT3 receptor EF = 17.73 q < .05
0.50 Oxymetazoline (descongestant, sympathomimetic) Molindone (indole deriv., antipsychotic) 0.77 Alpha-2-adrenergic receptor EF = 22.16 q < .05
0.65 Oxybuprocaine (local anesthetic) Metoclopramide (propulsive) 0.77 DNA repair protein RAD52 homolog EF = 6.57 q < .05
0.39 Niclosamide (salicylic acid deriv., anticestodal) Thalidomide (immunosuppressant) 0.76 Tyrosine-protein kinase SRC EF = 2.75 q < .05
0.41 Diethyltryptamine (psychedelic drug) Metoclopramide (propulsive) 0.75 5-HT6 receptor EF = 8.21 q < .05
0.35 Pentamidine (agent against Leishmaniasis/Trypanosomiasis) Haloperidol (antipsychotic, butyrophenone deriv.) 0.75 Muscarinic acetylcholine M4 EF = 11.22 q < .05
  1. Each drug-target association is predicted to cause different adverse effects confirmed in SIDER through the calculation of the EF and q-values [predicted adverse effects corroborated in SIDER (TP), predicted adverse effects not found in SIDER (FP), adverse effects described in SIDER and not predicted (FN), adverse effects not described in SIDER and not predicted by the model (TN)]
  2. aTC is the Tanimoto coefficient between both drugs using MACCS fingerprint
  3. bSimilar drug is the most similar drug binding the target in our ChEMBL data calculated with our 3D model
  4. c3D D-T is the 3D score that associates the drug candidate with the target according to our 3D model
  5. dEnrichment factor (EF) and q-values calculated for each drug-target association based on the integration of phenotype data from SIDER