/gtopdb-database

name: "gtopdb-database"
description: "Query IUPHAR/BPS Guide to Pharmacology (GtoPdb) for receptor-ligand interactions, target/ligand metadata, families, and approved drugs. Affinities (pKi/pIC50/pKd), action (Agonist/Antagonist/etc.), species, structures (SMILES/InChI). No auth. Always resolve targets via geneSymbol/accession; most metadata lives in sub-resources (/databaseLinks, /structure, /synonyms)."
license: "ODbL-1.0"

Guide to Pharmacology (GtoPdb) Database

Overview

The IUPHAR/BPS Guide to Pharmacology (GtoPdb) catalogues drug targets, ligands, and quantitative interactions across receptor pharmacology. The web services REST API at `https://www.guidetopharmacology.org/services/` returns JSON for targets, ligands, interactions, and family hierarchies. Base records are intentionally lean — gene symbols, UniProt accessions, ChEMBL IDs, SMILES/InChI all live in sub-resources (`/targets/{id}/databaseLinks`, `/targets/{id}/synonyms`, `/ligands/{id}/structure`, `/ligands/{id}/databaseLinks`). No authentication required.

When to Use

• Looking up the affinity (pKi/pIC50/pKd) of a ligand at a specific target
• Listing all annotated ligands for a receptor (e.g., μ-opioid receptor / OPRM1)
• Finding the approval status of a ligand (`approved=true`) and its cross-references (PubChem CID, ChEMBL ID, DrugBank ID)
• Retrieving the IUPHAR family hierarchy (867 families) for receptor classification
• Pulling structure descriptors (SMILES, InChI, InChIKey) for chemoinformatics
• Mapping HGNC symbol → UniProt → GtoPdb target ID for cross-database integration
• Use `chembl-database-bioactivity` for larger bioactivity datasets (2.4M+ compounds); GtoPdb is curated, smaller, with more annotation depth
• Use `dailymed-database` for FDA-approved drug labelling; GtoPdb is for pharmacology, not regulatory text

Prerequisites

• **Python packages**: `requests`, `pandas`, `matplotlib`
• **Data requirements**: HGNC symbols, UniProt accessions, GtoPdb target/ligand IDs, or drug INNs
• **Environment**: internet connection; no API key
• **Rate limits**: no published limits; use `time.sleep(0.2)` between requests in batch loops

pip install requests pandas matplotlib

Quick Start

import requests

BASE = "https://www.guidetopharmacology.org/services"

# Resolve HGNC symbol → GtoPdb target. geneSymbol= and accession= give an
# exact match. (name= matches across all fields and silently returns the
# wrong target — never use it for canonical lookups.)
r = requests.get(f"{BASE}/targets", params={"geneSymbol": "OPRM1"}, timeout=30)
targets = r.json()
print(f"OPRM1 hits: {len(targets)}")  # 1
t = targets[0]
print(f"targetId={t['targetId']}  name='{t['name']}'  type={t['type']}  family={t['familyIds']}")
# targetId=319  name='μ receptor'  type=GPCR  family=[50]

Core API

Query 1: Resolve a Target (HGNC symbol or UniProt accession)

import requests

BASE = "https://www.guidetopharmacology.org/services"

def find_target(*, geneSymbol=None, accession=None):
    """Exact-match target lookup. Pass ONE of geneSymbol or accession."""
    if geneSymbol:
        params = {"geneSymbol": geneSymbol}
    elif accession:
        params = {"accession": accession}
    else:
        raise ValueError("provide geneSymbol or accession")
    r = requests.get(f"{BASE}/targets", params=params, timeout=30)
    r.raise_for_status()
    hits = r.json()
    if not hits:
        return None
    return hits[0]

print(find_target(geneSymbol="OPRM1"))   # targetId 319 (μ receptor)
print(find_target(accession="P35372"))   # same — UniProt P35372 is OPRM1

# Base record has only IDs and family pointers — no gene symbol, UniProt, or
# synonym text. Read sub-resources to get those.
import requests
BASE = "https://www.guidetopharmacology.org/services"
print(requests.get(f"{BASE}/targets/319", timeout=30).json().keys())
# dict_keys(['targetId','name','type','familyIds','subunitIds','complexIds'])

Query 2: Target Cross-References and Synonyms

import requests, pandas as pd

BASE = "https://www.guidetopharmacology.org/services"

def target_xrefs(target_id):
    """Cross-database accessions: UniProt, HGNC, ChEMBL Target, Ensembl, etc."""
    r = requests.get(f"{BASE}/targets/{target_id}/databaseLinks", timeout=30)
    r.raise_for_status()
    return pd.DataFrame(r.json())

def target_synonyms(target_id):
    r = requests.get(f"{BASE}/targets/{target_id}/synonyms", timeout=30)
    r.raise_for_status()
    return [s.get("name") for s in r.json()]

df_links = target_xrefs(319)
print(df_links[["database", "accession", "species"]].head(8).to_string(index=False))
# database          accession    species
# ChEMBL Target     CHEMBL233    Human
# UniProtKB         P35372       Human
# HGNC              8156         Human
# ...
print("Synonyms:", target_synonyms(319))

Query 3: Target Interactions and Affinities

import requests, pandas as pd

BASE = "https://www.guidetopharmacology.org/services"

def target_interactions(target_id):
    """All ligand-target interaction records for a target.
    Each row carries ligandId, ligandName, type (Agonist/Antagonist/etc.),
    action, affinity (string), affinityParameter (pKi/pIC50/...), refs."""
    r = requests.get(f"{BASE}/targets/{target_id}/interactions", timeout=60)
    r.raise_for_status()
    rows = []
    for i in r.json():
        rows.append({
            "ligandId": i.get("ligandId"),
            "ligandName": i.get("ligandName"),
            "type": i.get("type"),                     # Agonist/Antagonist/Allosteric modulator/...
            "action": i.get("action"),
            "affinity": i.get("affinity"),             # string, may include "-" (range) or "~"
            "affinityParameter": i.get("affinityParameter"),  # pKi, pIC50, pKd, pEC50, pA2, pKB
            "species": i.get("targetSpecies"),
            "primary": i.get("primaryTarget"),
            "endogenous": i.get("endogenous"),