RSQM is a retrosynthesis tool designed to assist in planning synthesis paths for organic molecules. This tool utilizes multiple widely-used machine learning (ML) models for retrosynthesis planning. Each ML model is based on different algorithms and generates predictions in a unique way. By combining these different tools, RSQM aims to increase the possibility of generating a feasible synthetic path for a given input molecule.
- RSQM integrates multiple ML tools for retrosynthesis planning.
- Predictions made by RSQM are validated through two validation systems:
- Forward reaction validation system.
- Quantum mechanics validation system.
- RSQM provides an output of the top-rated synthesis paths based on the RS tools and validation systems.
- RS Adapter: The adapter module responsible for integrating various ML tools and managing their predictions.
- Forward Validation System: Module for validating predictions through forward reaction validation.
- QM Validation System: Module for validating predictions using quantum mechanics based tools.
This project is licensed under the MIT License.