the framework based on reinforcement learning for forward synthesis
Code for the paper "Synthetically Feasible De Novo Molecular Design of Leads Based on a Reinforcement Learning Model: AI-Assisted Discovery of an Anti-IBD Lead Targeting CXCR4"
This research is based on MolProphet: A One-Stop, General Purpose, and AI-Based Platform for the Early Stages of Drug Discovery
Website | Video Introduction | Paper
conda env create -f environment.yaml
conda activate rl_syn
You can download the processed data from this link
Our model checkpoints can be downloaded from GoogleDrive
Download and uncompress the model and processed data, then perform the following code
python demo.py
Jiang, X., Lu, L., Li, J., Jiang, J., Zhang, J., Zhou, S., Wen, H., Cai, H., Luo, X., Li, Z., Wang, J., Ju, B., & Bai, R. (2024). Synthetically Feasible De Novo Molecular Design of Leads Based on a Reinforcement Learning Model: AI-Assisted Discovery of an Anti-IBD Lead Targeting CXCR4. In Journal of Medicinal Chemistry. American Chemical Society (ACS). https://doi.org/10.1021/acs.jmedchem.4c00184
Yang, K., Xie, Z., Li, Z., Qian, X., Sun, N., He, T., Xu, Z., Jiang, J., Mei, Q., Wang, J., Qu, S., Xu, X., Chen, C., & Ju, B. (2024). MolProphet: A One-Stop, General Purpose, and AI-Based Platform for the Early Stages of Drug Discovery. In Journal of Chemical Information and Modeling (Vol. 64, Issue 8, pp. 2941–2947). American Chemical Society (ACS). https://doi.org/10.1021/acs.jcim.3c01979