FIRECODE is a computational chemistry workflow driver for the generation, optimization and refinement of conformational ensembles, also implementing some transition state utilities.
It implements flexible and customizable workflows for conformer generation (via CREST, RDKit), double-ended TS search (NEB via ASE, ML-FSM), and (constrained) ensemble optimization through popular calculators like XTB, TBLITE, ORCA, and Pytorch Neural Network models (AIMNET2, UMA) via ASE.
Conformational pruning is performed with the now standalone PRISM Pruner.
As a legacy feature from TSCoDe, FIRECODE can also assemble non-covalent adducts from conformational ensembles (embedding) programmatically.
The package is distributed via pip, and the use of uv is highly recommended. The default installation is minimalistic, and torch/GPU support requires dedicated installs:
uv pip install firecode # XTB, TBLITE, ORCA
uv pip install firecode[aimnet2] # + AIMNET2
uv pip install firecode[uma] # + UMA/OMOL
uv pip install firecode[full] # + AIMNET2, UMA/OMOL
More installation details in the documentation.
Additional documentation on how to install and use the program can be found on readthedocs.