Table 2. Key databases for pharmacological research.

There are several databases that artificial intelligence (AI) leverages to aid in drug discovery and precision medicine. A few are mentioned here.

TCGA: The Cancer Genome Atlas; ADMET: absorption, distribution, metabolism, excretion, and toxicity

Database	Description	Used for
LinkedOmics	Comprehensive database of cancer clinical and molecular data. It gathers TCGA cancer-related multi-omics, clinical, and mass spectrometry proteomics data.	Target identification
DepMap portal	Website portal offering analytical and visualization tools for cancer. It includes cancer cell line sensitivity and genetic data.	Target identification
Therapeutic target database	Database of linked medications and recognized therapeutic proteins, nucleic acids, and diseases.	Target identification
DUD-E	Programs for benchmarking molecular docking by providing challenging decoy.	Hit identification
CSAR	Benchmark databases of protein-ligand complexes with various crystal structures and binding affinities.	Hit identification
BindingDB	An online database of measured binding affinities focused primarily on the interaction of drug target proteins with small drug-like molecules.	Hit identification, ADMET property prediction
DrugBank	Free comprehensive database of drugs and drug targets. It contains different chemicals and target information for each drug.	Hit identification, ADMET property prediction, training deep generative models
MATADOR	Integrated medication information on medical indications, adverse drug effects, drug metabolism, target protein pathways, and gene ontology terms.	Hit identification
PubChem	Integrated chemistry database. It includes small to large molecules with structure, physical properties, bioactivity, patents, etc.	Hit identification, ADMET property prediction, training deep generative models
ChemIDplus	An online search portal that provides access to chemicals listed in the National Library of Medicine databases	ADMET property prediction
ToxRefDB	Data were collected from more than 5000 in vivo toxicity studies, to contain 10 types of toxicity studies.	ADMET property prediction
GDB-13	A fully cataloged virtual database based on simple chemical stability and synthetic feasibility, up to 13 atoms C, N, O, S, and Cl.	Training deep generative models
GDB-17	Fully cataloged virtual database based on simple chemical stability and synthetic feasibility, up to 17 C, N, O, S, and halogen atoms	Training deep generative models
ChEMBL	A curated database of bioactive drug-like small molecules. It mainly covers 2D structures, calculated properties, and bioactivity.	Hit identification, ADMET property prediction, training deep generative models