1.Automated extraction of chemical synthesis actions from experimental procedures
A Python library with the action definition and handling as well as associated scripts for training the transformer model can be found on GitHub at https://github.com/rxn4chemistry/paragraph2actions. The trained models can be freely used online at https://rxn.res.ibm.com or with the Python wrapper at https://github.com/rxn4chemistry/rxn4chemistry to extract action sequences from experimental procedures.
2.Computational planning of the synthesis of complex natural products
https://www.nature.com/articles/s41586-020-2855-y#Sec21