ChEMBL on the platform

ChEMBL lets you look up measured bioactivity — which compounds have been tested against a target and how potent they were (IC50, Ki, EC50, Kd), curated from medicinal chemistry literature. Reach for it first when you already know the target and want known actives to seed SAR. When you instead want to search by structure, similarity, or computed properties, use its sibling PubChem; ChEMBL provides measured data, PubChem provides broader compound coverage. See what you can achieve.

Try it

ChEMBL on the platform

---

ChEMBL is EMBL-EBI's open bioactivity database of distinct compounds, targets, and activity measurements curated from medicinal chemistry literature. The platform integrates ChEMBL with four specialized search modes, calling the ChEMBL REST API directly (https://www.ebi.ac.uk/chembl/api/data) rather than the chembl_webresource_client Python package.

ChEMBL data is extracted from peer-reviewed publications by a team of curators at EMBL-EBI, making it one of the most reliable public sources for structure-activity relationship (SAR) data. Each record is linked to its source publication and assigned a confidence score reflecting how directly the target was identified in the original assay.

Citations

Zdrazil B, Felix E, Hunter F, Manners EJ, Blackshaw J, Corbett S, et al. The ChEMBL Database in 2023: a drug discovery platform spanning multiple bioactivity data types and time periods. Nucleic Acids Research. 2024;52(D1):D1180-D1192. doi: 10.1093/nar/gkad1004

Four Search Modes

When to Use ChEMBL

Primary applications

• Hit expansion: Find known actives against your target to seed SAR analysis
• Competitive intelligence: See which compounds have been tested against a target and their potency ranges
• Scaffold hopping: Use similarity search to find analogs with different core scaffolds
• Target identification: Look up which targets a compound (or similar compounds) are known to bind
• Assay data aggregation: Collect IC50, Ki, EC50, and Kd data across multiple studies for a compound-target pair

When to use other tools instead

• Compound properties only (no bioactivity): Use PubChem — broader compound coverage with computed properties
• Clinical/regulatory status: Use DrugCentral or the regulatory tools for approval and safety data
• Protein structure and function: Use UniProt for sequence annotation and PDB for 3D structures
• Predicted ADMET: Use ADMET-AI or ADMETLab3 — ChEMBL provides measured data, not predictions

Key Metrics Explained

Predefined Examples

• EGFR (CHEMBL203) IC50 <100 nM
• ABL1 (CHEMBL1862) IC50 <50 nM
• Aspirin similarity 0.8
• Phenyl ring substructure search
• Single-protein human targets related to BRAF

Chat Usage

Example Prompts

• "Search ChEMBL for EGFR inhibitors with IC50 below 100 nM"
• "Find compounds similar to imatinib in ChEMBL"
• "What is the ChEMBL target ID for BRAF?"
• "Show me all kinase inhibitors with Ki < 10 nM"

Workflow Chaining

• Feeds into: ADMET-AI (drug-likeness screening), DiffDock (molecular docking), Compound Passport (full compound profile)
• Accepts from: PubChem (compound SMILES), UniProt (target identifiers), PubMed (target names mentioned in retrieved abstracts)

FAQ

Read next