Database Search
Syndirella offers three options for database search functionality: Arthor, Hippo, and Postera (Manifold). All three tools support superstructure searches to find commercially available compounds that can be used as reactants in synthesis.
Arthor
Arthor is a free, publicly accessible database search tool that provides access to multiple chemical databases including ZINC, Enamine, Mcule, and stock databases.
Usage:
When running Syndirella, specify Arthor as the database search tool:
syndirella run --input input.csv --output output_dir --db_search_tool arthor
Default Vendors:
The default vendor list includes:
- enamine_real - Enamine REAL Database
- stock - In-Stock database
- mcule - Mcule databases (includes multiple Mcule variants)
- zinc - ZINC databases (includes multiple ZINC variants)
Available Vendor Options:
Arthor supports a wide range of vendor databases: check the Arthor documentation for more details.
Checking Available Databases:
You can programmatically check which databases are available:
from syndirella.database.Arthor import Arthor
arthor = Arthor()
available_dbs = arthor.get_available_databases()
print(available_dbs)
Postera (Manifold)
Postera’s Manifold API provides access to a comprehensive database of commercially available compounds. It supports superstructure searches across multiple vendor catalogs.
Setup:
Get your Manifold API key from your Postera account.
Set the required environment variables:
export MANIFOLD_API_KEY=[YOUR_API_KEY]
export MANIFOLD_API_URL=[API_URL]
# Example:
# export MANIFOLD_API_KEY=your_api_key_here
# export MANIFOLD_API_URL=https://api.postera.ai
Attention
A Manifold Pro account is required for API access.
Usage:
When running Syndirella, specify Postera as the database search tool:
syndirella run --input input.csv --output output_dir --db_search_tool manifold
Default Vendors:
The default vendor list includes:
- enamine_bb - Enamine Building Blocks
- mcule - Mcule database
- mcule_ultimate - Mcule Ultimate database
- enamine_real - Enamine REAL database
- enamine_made - Enamine Made database
Hippo
Hippo provides local database search functionality using a HIPPO database file. This option is useful when you have your own database or want to search locally without API dependencies.
Installation
HIPPO requires rdkit>=2024.9.6; AiZynthFinder declares rdkit<2024. Install in this order so pip does not pull AiZynthFinder’s RDKit:
pip install aizynthfinder --no-deps
pip install syndirella[hippo]
conda install -c conda-forge chemicalite
See the HIPPO documentation for full setup and version compatibility. To create a minimal example database (e.g. at ~/CursorProjects/HIPPO/reference_db.sqlite), run python scripts/create_hippo_example_db.py [OUTPUT_DIR].
Setup
Have a HIPPO database file available and initialize the Hippo search with the database path.
Usage
Specify Hippo as the database search tool:
syndirella run --input input.csv --output output_dir --db_search_tool hippo --reference_db /path/to/database
Default Behavior
If no database search tool is specified, Syndirella will use the default tool (Arthor).