README.md

SynFlowNet - Design of Diverse and Novel Molecules with Synthesis Constraints

Official implementation of SynFlowNet, a GFlowNet model with a synthesis action space. The paper is available on arxiv.

Primer

SynFlowNet is a GFlowNet model that generates molecules from chemical reactions and available building blocks. SynFlowNet is trained to sample molecules with probabilities proportional to their rewards. This repo contains instructions for how to train SynFlowNet and sample synthesisable molecules. The code builds upon the codebase provided by recursionpharma/gflownet, available under the MIT license. For a primer and repo overview visit recursionpharma/gflownet.

Installation

PIP

This package is installable as a PIP package, but since it depends on some torch-geometric package wheels, the --find-links arguments must be specified as well:

pip install -e . --find-links https://data.pyg.org/whl/torch-2.1.2+cu121.html

Or for CPU use:

pip install -e . --find-links https://data.pyg.org/whl/torch-2.1.2+cpu.html

Getting started

Data

The training relies on two data sources: modified Hartenfeller-Button reaction templates and Enamine building blocks. The building blocks are not freely available and can be obtained upon request from enamine.net/building-blocks/building-blocks-catalog. Instructions can be found in src/gflownet/data/building_blocks/.

Training

The model can be trained by running src/gflownet/tasks/reactions_task.py using different reward functions implemented in the same file. You may want to change the default configuration in main().

[Optional] If using GPU-accelerated Vina

For easy adoption to other targets, a GPU-accelerated version of Vina docking can be used to calculate rewards as binding affinities to targets of interest. Follow the instructions at this repo to compile an excuteable for QuickVina2-GPU-2-1. One done, place the excuteable in bin/.

Citation

If you use this code in your research, please cite the following paper:

@article{cretu2024synflownet,
      title={SynFlowNet: Design of Diverse and Novel Molecules with Synthesis Constraints},
      author={Miruna Cretu, Charles Harris, Ilia Igashov, Arne Schneuing, Marwin Segler, Bruno Correia, Julien Roy, Emmanuel Bengio and Pietro Liò},
      journal={arXiv preprint arXiv},
      year={2024}
}