SynPlanner: An End-to-End Tool for Synthesis Planning

SynPlanner is an\nend-to-end tool for designing customized retrosynthetic\nplanners from reaction data. It includes a reaction data curation\npipeline (reaction atom-to-atom mapping, reaction standardization,\nand filtration), reaction rule extraction, retrosynthetic model training,\nand retrosynthetic planning. The tool is designed to be as flexible\nas possible, supporting the customization of each step of the pipeline\nto address different needs in the development of customized retrosynthetic\nplanning solutions. The retrosynthetic planning in SynPlanner is performed\nby Monte Carlo Tree Search (MCTS) guided by graph neural networks\nfor node expansion (retrosynthetic rule predictions) and evaluation\n(precursor synthesizability prediction). The solution can be accessed\nby a simple graphical user interface and a command line interface\nand is accompanied by a collection of tutorials. SynPlanner is available\non GitHub at https://github.com/Laboratoire-de-Chemoinformatique/SynPlanner.