Supporting code to the Publication Integrating Reaction Schemes, Reagent Databases, and Virtual Libraries into Fragment-Based Design by Reinforcement Learning; doi.org/10.1021/acs.jcim.3c00735

Installation

Clone Repository

git clone https://github.com/Sanofi-Public/IDD-papers-fragrl.git
cd IDD-papers-fragrl
git submodule update --init  ## to get MolScore submodule

Create and activate conda environment

conda env create -f environment.yml
conda activate env_fragrl

Usage

python /path/to/deepfmpo/Main.py -c configuration.txt

or (for DeepVL):

python /path/to/deepfmpo/DeepVL.py -c configuration.txt

For more details see folder documentation.

Contact

christoph.grebner@sanofi.com