aizynthfinder is a tool for retrosynthetic planning. The algorithm is based on a Monte Carlo tree search that recursively breaks down a molecule to purchasable precursors. The tree search is guided by a policy that suggests possible precursors by utilizing a neural network trained on a library of known reaction templates.
Before you begin, ensure you have met the following requirements:
- Linux, Windows or Mac platforms are supported - as long as the dependencies are supported on these platforms.
The tool has been developed on a Linux platform, but the software has been tested on Windows 10 and macOS Catalina.
- You have installed anaconda or miniconda with python 3.6 or 3.7
To install aizynthfinder, follow these steps:
- First, install these conda packages
conda install -c rdkit "rdkit=>2019.09.1" -y
conda install -c anaconda tensorflow>=2.1.0 -y
conda install graphviz -y
if you have GPU and CUDA libraries enabled on your machine, you can install the tensorflow-gpu
package instead.
- Secondly, install the
aizynthfinder
package
python -m pip install https://github.com/MolecularAI/aizynthfinder/archive/v1.0.0.tar.gz
if you want to install the latest version
or
python -m pip install -e .
if you are a developer, using the repository.
Note on the graphviz installation: this package does not depend on any third-party python interfaces to graphviz but instead calls the neato
and dot
executables directly. If these executable are not in the $PATH
environmental variable, the generation of route images will not work. If unable to install it properly with the default conda chanel, try using -c anaconda
.
The tool will install the aizynthcli
and aizynthapp
tools
as interfaces to the algorithm:
aizynthcli --config config.yml --smiles smiles.txt
aizynthapp --config config.yml
Consult the documentation here for more information.
To use the tool you need
1. A stock file
2. A trained rollout policy network (including the Keras model and the list of unique templates)
Such files can be downloaded from figshare or they can be downloaded automatically using
download_public_data my_folder
where my_folder
is the folder that you want download to.
This will create a config.yml
file that you can use with either aizynthcli
or aizynthapp
.
Tests uses the pytest
package.
To use, first install the dependencies
python -m pip install -r requirements_dev.txt
and then run the tests using
pytest -v
We welcome contributions, in the form of issues or pull requests.
If you have a question or want to report a bug, please submit an issue.
To contribute with code to the project, follow these steps:
- Fork this repository.
- Create a branch:
git checkout -b <branch_name>
. - Make your changes and commit them:
git commit -m '<commit_message>'
- Push to the remote branch:
git push
- Create the pull request.
Please use black
package for formatting, and follow pep8
style guide.
The contributors have limited time for support questions, but pleae do not hesitate to submit an issue (see above).
The software is licensed under the MIT license (see LICENSE file), and is free and provided as-is.
- Thakkar A, Kogej T, Reymond J-L, et al (2019) Datasets and their influence on the development of computer assisted synthesis planning tools in the pharmaceutical domain. Chem Sci. https://doi.org/10.1039/C9SC04944D
- Genheden S, Thakkar A, Chadimova V, et al (2020) AiZynthFinder: a fast, robust and flexible open-source software for retrosynthetic planning. ChemRxiv. Preprint. https://doi.org/10.26434/chemrxiv.12465371.v1