Pubchem Compound Search

Prerequisites

• Python packages: pubchempy, requests (for direct API), pandas (for batch processing)
• No API key required: PubChem is freely accessible
• Rate limits: Max 5 requests/second, 400 requests/minute

pip install pubchempy requests pandas

Quick Start

import pubchempy as pcp

# Search by name → get properties
compound = pcp.get_compounds("aspirin", "name")[0]
print(f"CID: {compound.cid}")
print(f"SMILES: {compound.canonical_smiles}")
print(f"MW: {compound.molecular_weight}, LogP: {compound.xlogp}")
print(f"HBD: {compound.h_bond_donor_count}, HBA: {compound.h_bond_acceptor_count}")

Workflow

Step 1: Compound Search

Search by name, CID, SMILES, InChI, or molecular formula.

import pubchempy as pcp

# By name
compounds = pcp.get_compounds("caffeine", "name")
print(f"Found {len(compounds)} compounds for 'caffeine'")

# By CID (fastest)
compound = pcp.Compound.from_cid(2244)  # Aspirin
print(f"CID 2244 = {compound.iupac_name}")

# By SMILES
compound = pcp.get_compounds("CC(=O)OC1=CC=CC=C1C(=O)O", "smiles")[0]
print(f"SMILES lookup: CID {compound.cid}")

# By molecular formula (returns all matches)
formula_matches = pcp.get_compounds("C9H8O4", "formula")
print(f"Formula C9H8O4 matches: {len(formula_matches)} compounds")

Step 2: Property Retrieval

Get molecular properties for one or more compounds.

import pubchempy as pcp

# Full compound object
compound = pcp.get_compounds("ibuprofen", "name")[0]
print(f"MW: {compound.molecular_weight}")
print(f"LogP: {compound.xlogp}")
print(f"TPSA: {compound.tpsa}")
print(f"Rotatable bonds: {compound.rotatable_bond_count}")

# Selective property retrieval (more efficient for specific needs)
props = pcp.get_properties(
    ["MolecularWeight", "XLogP", "TPSA", "HBondDonorCount"],
    "aspirin", "name"
)
print(props)  # List of dicts

Step 3: Similarity Search

Find structurally similar compounds using Tanimoto coefficient.

import pubchempy as pcp

# Get reference compound SMILES
ref = pcp.get_compounds("gefitinib", "name")[0]

# Similarity search (may take 15-30s for async processing)
similar = pcp.get_compounds(
    ref.canonical_smiles, "smiles",
    searchtype="similarity",
    Threshold=85,       # Tanimoto threshold (0-100)
    MaxRecords=50
)
print(f"Found {len(similar)} compounds with ≥85% similarity to gefitinib")
for comp in similar[:5]:
    print(f"  CID {comp.cid}: MW={comp.molecular_weight}")

Step 4: Substructure Search

Find compounds containing a specific structural motif.

import pubchempy as pcp

# Search for sulfonamide-containing compounds
hits = pcp.get_compounds(
    "S(=O)(=O)N", "smiles",
    searchtype="substructure",
    MaxRecords=100
)
print(f"Found {len(hits)} compounds with sulfonamide group")

Step 5: Bioactivity Data Access

Retrieve biological screening results via PUG-REST API.

import requests

cid = 2244  # Aspirin
url = f"https://pubchem.ncbi.nlm.nih.gov/rest/pug/compound/cid/{cid}/assaysummary/JSON"
response = requests.get(url)

if response.status_code == 200:
    data = response.json()
    rows = data.get("Table", {}).get("Row", [])
    print(f"Aspirin has {len(rows)} bioassay records")

Step 6: Batch Property Comparison

Compare properties across multiple compounds.

import pubchempy as pcp
import pandas as pd
import time

compounds = ["aspirin", "ibuprofen", "naproxen", "celecoxib"]
results = []

for name in compounds:
    comp = pcp.get_compounds(name, "name")[0]
    results.append({
        "Name": name, "CID": comp.cid,
        "MW": comp.molecular_weight, "LogP": comp.xlogp,
        "TPSA": comp.tpsa, "HBD": comp.h_bond_donor_count,
        "HBA": comp.h_bond_acceptor_count,
    })
    time.sleep(0.25)  # Respect rate limits

df = pd.DataFrame(results)
print(df.to_string(index=False))

Step 7: Identifier Format Conversion

Convert between chemical identifier formats.

import pubchempy as pcp

compound = pcp.get_compounds("caffeine", "name")[0]
print(f"CID:      {compound.cid}")
print(f"IUPAC:    {compound.iupac_name}")
print(f"SMILES:   {compound.canonical_smiles}")
print(f"InChI:    {compound.inchi}")
print(f"InChIKey: {compound.inchikey}")
print(f"Formula:  {compound.molecular_formula}")

# Download structure files
pcp.download("SDF", "caffeine", "name", "caffeine.sdf", overwrite=True)
print("Downloaded caffeine.sdf")

Key Parameters

Common Recipes

Recipe: Drug-Likeness Screening (Lipinski's Rule of Five)

When to use: Quick check if a compound is orally bioavailable.

import pubchempy as pcp

def check_lipinski(name):
    comp = pcp.get_compounds(name, "name")[0]
    rules = {
        "MW ≤ 500": comp.molecular_weight <= 500,
        "LogP ≤ 5": (comp.xlogp or 0) <= 5,
        "HBD ≤ 5": comp.h_bond_donor_count <= 5,
        "HBA ≤ 10": comp.h_bond_acceptor_count <= 10,
    }
    violations = sum(1 for v in rules.values() if not v)
    return rules, violations

rules, v = check_lipinski("metformin")
print(f"Violations: {v}/4 — {'PASS' if v <= 1 else 'FAIL'}")
for rule, passed in rules.items():
    print(f"  {'✓' if passed else '✗'} {rule}")

Recipe: Get All Synonyms for a Compound

When to use: Finding alternative names, trade names, or CAS numbers.

import pubchempy as pcp

synonyms = pcp.get_synonyms("aspirin", "name")
if synonyms:
    names = synonyms[0]["Synonym"]
    print(f"Found {len(names)} synonyms for aspirin:")
    for name in names[:10]:
        print(f"  {name}")

Recipe: Download 2D Structure Image

When to use: Generating structure images for reports or presentations.

import requests

cid = 2519  # Caffeine
url = f"https://pubchem.ncbi.nlm.nih.gov/rest/pug/compound/cid/{cid}/PNG?image_size=large"
response = requests.get(url)
with open("caffeine_structure.png", "wb") as f:
    f.write(response.content)
print("Saved caffeine_structure.png")

Expected Outputs

• Compound search: pubchempy.Compound objects with properties (CID, name, SMILES, MW, etc.)
• Property retrieval: List of dictionaries with requested properties
• Similarity search: List of Compound objects sorted by similarity
• Bioactivity query: JSON with assay results (activity outcome, assay ID, target)
• Structure download: SDF, JSON, or PNG files

Troubleshooting

References

• PubChem PUG-REST API — official REST API docs
• PubChemPy documentation — Python wrapper docs
• PubChem PUG-REST tutorial — step-by-step guide
• PubChem database — web interface