Retrosynthesis route finder — AiZynthFinder + Rxn-INSIGHT + Chemistry by Design
Project description
Path Finder
AiZynthFinder · Chemistry by Design · Rxn-INSIGHT
Yara Chahda · Corentin Postmann · Inès Ouchen — EPFL 2025
Retrosynthesis route finder: find, rank and compare synthesis routes for any target molecule.
Installation
Step 1 — Install RDKit (conda only)
conda install -c conda-forge rdkit
RDKit cannot be installed via pip — this single conda step is required.
Step 2 — Install Path Finder
pip install path-finder-retrosynthesis
With predicted routes (Rxn-INSIGHT):
pip install path-finder-retrosynthesis[predicted]
Step 3 — First-time setup
path-finder-setup
The wizard copies the bundled datasets, creates data/config.yml, and tells you exactly which files to download.
Step 4 — Download AiZynthFinder model files
Go to https://github.com/MolecularAI/aizynthfinder/releases and download:
| File | Purpose |
|---|---|
uspto_model.onnx |
Expansion policy network |
uspto_templates.csv.gz |
Reaction templates |
uspto_filter_model.onnx |
Filter policy (optional) |
zinc_stock.hdf5 |
Purchasable building blocks |
Place them in data/aizynthfinder/.
Step 5 — Edit config.yml
Open data/config.yml and replace /PATH/TO/AIZYNTHFINDER/ with the absolute path to your model files.
macOS / Linux:
expansion:
uspto:
- /Users/alice/data/aizynthfinder/uspto_model.onnx
- /Users/alice/data/aizynthfinder/uspto_templates.csv.gz
Windows:
expansion:
uspto:
- C:/Users/alice/data/aizynthfinder/uspto_model.onnx
- C:/Users/alice/data/aizynthfinder/uspto_templates.csv.gz
Step 6 — Launch
path-finder
Open http://localhost:8501 in your browser.
Quick summary
conda install -c conda-forge rdkit
pip install path-finder-retrosynthesis
path-finder-setup
# edit data/config.yml
path-finder
Data files
| File | Bundled | Description |
|---|---|---|
reaction_dataset.json |
✅ | Curated synthesis routes |
toxicity_dataset.json |
✅ | Safety scores |
generic_reactions.json |
✅ | 10 000 USPTO reactions |
config.yml |
❌ | Created by setup wizard |
data/aizynthfinder/ |
❌ | Model files — download separately |
uspto_rxn_insight.gzip |
❌ | Rxn-INSIGHT database (optional) |
Troubleshooting
| Problem | Solution |
|---|---|
config.yml not found |
Run path-finder-setup |
| AiZynthFinder crash | Check paths in config.yml are absolute |
No routes found |
Try Galanthamine or Morphine |
| Rxn-INSIGHT disabled | pip install path-finder-retrosynthesis[predicted] |
| Slow search (~2 min) | Normal — AiZynthFinder MCTS is intensive |
Developer setup
git clone https://github.com/YaraChahda/path_finder.git
cd path_finder
conda install -c conda-forge rdkit
pip install -e ".[predicted]"
path-finder-setup
path-finder
Citation
- AiZynthFinder: Genheden et al., J. Cheminf. 2020
- Rxn-INSIGHT: Thakkar et al., J. Cheminf. 2023
- Open Reaction Database: Kearnes et al., JACS 2021
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Filter files by name, interpreter, ABI, and platform.
If you're not sure about the file name format, learn more about wheel file names.
Copy a direct link to the current filters
File details
Details for the file path_finder_retrosynthesis-1.0.2.tar.gz.
File metadata
- Download URL: path_finder_retrosynthesis-1.0.2.tar.gz
- Upload date:
- Size: 1.0 MB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/6.2.0 CPython/3.13.9
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
9f0c104a9c500d3e31778ca7f32beadc3822867d943ff2c41394a288c7bb172e
|
|
| MD5 |
5a3127c00131277138e895fceaeb273b
|
|
| BLAKE2b-256 |
9caaaaaa29f722403ea0eb02331cbd4889aceab469419ca80e5f79e2c432a7ab
|
File details
Details for the file path_finder_retrosynthesis-1.0.2-py3-none-any.whl.
File metadata
- Download URL: path_finder_retrosynthesis-1.0.2-py3-none-any.whl
- Upload date:
- Size: 1.0 MB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/6.2.0 CPython/3.13.9
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
58512716123f4d201449d8780fd559b31e1eeec874343932f4a67338a484e01e
|
|
| MD5 |
6ddbffe8452676c982eff070416f23c4
|
|
| BLAKE2b-256 |
893d2934b38d6d59d4bb89163271a5cb79a44ed71cec1c1285c174c82ebe7ff5
|