A python package prototype for the exploration of enzymatic mechanisms. The code is published as part of the supplementary material to the submitted paper "Graph Transformation for Enzymatic Mechanisms".
We provide a demo version of our approach online which can be found at https://cheminf.imada.sdu.dk/mechsearch.
MechSearch depends on NetworkX and a custom version of MØD which can be found here.
The dependencies can be installed as conda packages using the following command:
conda create -n mechsearch -c jakobandersen/label/ECCB-21 -c jakobandersen -c conda-forge networkx mod=0.12.0.19
conda activate mechsearch
Additionally, a docker image of the custom version of MØD is available:
docker pull jakobandersen/mod:ECCB-21
To test if the dependencies were installed correctly a small example
is provided in the file scripts/square.py. Said script also serves
as a simple example to see how state spaces are created and
pathways enumerated. To run the script from the root folder type:
mod -f scripts/square.py
A summary file will be located at summary/summary.pdf.
The state spaces of RHEA reactions that were analysed in the submitted paper can be constructed by running the following command:
mod -f scripts/rhea_analysis/compute_state_spaces.py
The state spaces will by default be stored in the folders
state_spaces_fixed_amino and state_spaces_1_amino.
To generate pathways of a given set of state spaces, we run the following command:
mod -f scripts/rhea_analysis/print_paths.py
For more information about how each script functions, please see the docstrings located in both scripts.