SmileSherlock

A high-performance, production-grade tool for SMILES validation, PubChem lookup, and chemical structure retrieval.

Features

• SMILES Validation & Canonicalization - Validate and standardize SMILES strings using RDKit
• Multi-format Input - Support for CSV, TSV, XLSX, SMI, SDF, and TXT files
• Smart Auto-detection - Automatically identify SMILES columns
• PubChem Lookup - Search by SMILES, CID, Name, InChI, and InChIKey
• Rich Metadata - Retrieve IUPAC name, molecular formula, mass, descriptors
• Structure Downloads - Get 2D/3D SDF, MOL, PDB, and PNG formats from PubChem
• Offline Molecule Generation (--gen) - Generate 2D and 3D conformations (SDF, MOL, PDB) offline from SMILES via RDKit with forcefield optimization
• Molecular Fingerprints - ECFP4, ECFP6, FCFP4, MACCS, RDKit, AtomPair, Torsion fingerprints offline (fingerprint command)
• Similarity Search - Tanimoto-based library search with threshold and top-N ranking (similar command)
• Drug-Likeness Filtering - Lipinski Ro5, Veber, Ghose, Egan, Ro3, PAINS alerts, and QED scoring (filter command)
• Batch Processing - Process hundreds of compounds with progress tracking
• Async/Multithreading - Fast parallel downloads with retry logic
• Caching - SQLite database for storing results locally
• Multiple Exports - Save results as CSV, Excel, or JSON
• Python API - Use directly in your scripts via smilesherlock module
• CLI Tool - Full-featured command-line interface with smilesherlock command

Installation

From PyPI

pip install smilesherlock

Development Installation

Clone the repository and install in editable mode:

git clone https://github.com/AtharvaTilewale/SmileSherlock.git
cd SmileSherlock
pip install -e ".[dev]"

Quick Start

CLI Usage

# Show configuration and status
smilesherlock status

# Initialize directories and database
smilesherlock init

# Lookup a single compound (by SMILES, CID, or Chemical Name)
smilesherlock lookup "c1ccccc1"  # Benzene
smilesherlock lookup "aspirin"
smilesherlock lookup 5282253 --type cid

# Batch process a file to retrieve metadata
smilesherlock batch compounds.csv --output results.xlsx --format xlsx

# Download structure from PubChem
smilesherlock download 5282253 --format sdf --3d

# Generate 3D structure offline from SMILES using RDKit (--gen all)
smilesherlock download "CC(=O)OC1=CC=CC=C1C(=O)O" --gen all --3d --format sdf

# Generate 2D MOL structure locally from SMILES
smilesherlock download "c1ccccc1" --gen all --2d --format mol

# Batch download with offline fallback for missing structures (--gen missing)
smilesherlock download --file compounds.csv --gen missing --3d --format sdf --output-dir ./structures/

# Batch generate all structures offline from a SMILES file (--gen all)
smilesherlock download --file compounds.smi --gen all --3d --format pdb --output-dir ./3d_models/

Molecular Fingerprints

# Generate ECFP4 fingerprint (single compound)
smilesherlock fingerprint "CC(=O)OC1=CC=CC=C1C(=O)O" --type ecfp4

# All 7 fingerprint types at once
smilesherlock fingerprint "CC(=O)OC1=CC=CC=C1C(=O)O" --type all

# Batch - save to CSV
smilesherlock fingerprint --file compounds.smi --type maccs --output fingerprints.csv

Similarity Search

# Top-10 most similar compounds (Tanimoto >= 0.5)
smilesherlock similar "CC(=O)OC1=CC=CC=C1C(=O)O" --file library.smi --threshold 0.5 --top 10

# Save hits to CSV
smilesherlock similar "CCO" --file compounds.csv --fp-type ecfp6 --output hits.csv

Drug-Likeness Filtering

# Single compound - all rules
smilesherlock filter "CC(=O)OC1=CC=CC=C1C(=O)O"

# Batch - keep only Lipinski-compliant, PAINS-free compounds
smilesherlock filter --file compounds.csv --rules lipinski,veber,pains --output drug_like.csv

# Remove PAINS compounds from a library
smilesherlock filter --file library.csv --rules pains --output no_pains.csv

# Lead-like compounds with minimum QED 0.5
smilesherlock filter --file library.csv --rules ro3 --qed-min 0.5 --output leads.csv

Python API

from smilesherlock import lookup, lookup_file, download_structure, generate_structure, validate_smiles

# Lookup single compound
result = lookup("c1ccccc1")
print(result.cid, result.iupac_name)

# Process batch file
results = lookup_file("compounds.csv", output_format="xlsx")

# Download structure from PubChem
download_structure(5282253, format="sdf", dimension="3d")

# Generate 2D or 3D structure offline from SMILES
generate_structure(
    smiles="CC(=O)OC1=CC=CC=C1C(=O)O",
    output_path="aspirin_3d.sdf",
    format="sdf",
    dimension="3d",
    title="Aspirin"
)

For more detailed API documentation, see the API Reference page.

Documentation

For complete tutorials and advanced usage examples, see the Practical Guide or visit the official documentation on Read the Docs.

Requirements

• Python 3.10+
• RDKit (cheminformatics library)
• pandas (data handling)
• requests/aiohttp (HTTP)
• typer (CLI framework)
• rich/tqdm (UI/progress)

Configuration

For configuration and architecture details, see the Configuration & Architecture page.

Contributing

Contributions are welcome! Please:

1. Fork the repository
2. Create a feature branch (git checkout -b feature/amazing-feature)
3. Commit changes (git commit -m 'Add amazing feature')
4. Push to branch (git push origin feature/amazing-feature)
5. Open a Pull Request

For more details, see the Contributing Guide.

License

This project is licensed under the MIT License - see the LICENSE file for details.

Citation

If you use SmileSherlock in your research, please cite:

@software{smilesherlock2026,
  author={Atharva Tilewale},
  doi={10.5281/zenodo.21763825},
  month={8},
  title={SmileSherlock: A High-Performance SMILES Validation and PubChem Lookup Tool},
  version={1.2.0},
  year={2026},
  url={https://github.com/AtharvaTilewale/SmileSherlock}
}

Support

• Documentation: https://smilesherlock.readthedocs.io
• Issues: https://github.com/AtharvaTilewale/SmileSherlock/issues
• Discussions: https://github.com/AtharvaTilewale/SmileSherlock/discussions

Changelog

See CHANGELOG.md for version history.

Made with ❤️ for the cheminformatics community