From: eToxPred: a machine learning-based approach to estimate the toxicity of drug candidates
Dataset | Size | Usage | Description |
---|---|---|---|
NuBBE | 1008 | Train/test (SAscore) | Natural products and derivatives from the Brazilian biodiversity |
UNPD | 81,372 | Train/test (SAscore) | Diverse collection of natural products |
DUD-E (actives) | 17,499 | Train/test (SAscore) | Mostly synthetic bioactive compounds against 102 protein targets |
FDA-approved | 1515 | Train/test (SAscore) Train (Tox-score) | FDA approved drugs from DrugBank |
KEGG-Drug | 3682 | Test (Tox-score) | Drugs approved in Japan, United States, and Europe |
TOXNET | 3035 | Train (Tox-score) | Potentially hazardous chemicals |
T3DB | 1283 | Test (Tox-score) | Collection of pollutants, pesticides, drugs, and food toxins |
TCM | 5883 | Test (SAscore, Tox-score, unlabeled) | Traditional Chinese medicines |
CP | 1401 | Train/test (specific toxicity) | Carcinogenic compounds tested in rodents |
CD | 1571 | Train/test (specific toxicity) | Cardiotoxic compounds tested against hERG potassium channel |
ED | 17,059 | Train/test (specific toxicity) | Endocrine disrupting compounds tested against androgen and estrogen receptors |
AO | 12,612 | Train/test (specific toxicity) | Toxins from various sources annotated with acute oral toxicity |