A pipeline to help identify and classify structures in tomography data using global and local topological (network) features.
The package consists of 3 core modules:
pdb2graph.py
- converts a PDB structure into a graph network.density2graph.py
- converts a 3D density to a graph.graph2class.py
- measures graph features and classifies
- open terminal / command prompt
- clone the repository:
git clone https://github.com/EMSL-Computing/grip-tomo
- install the dependencies:
pip install -r /path/to/requirements.txt
- run tests:
cd path/to/grip_tomo/
thenpython _test_basic.py
- review example notebook,
grip_tomo/example_notebook.ipynb
- To interact with
.ipynb
files install Jupyter-lab
- To interact with
Please cite this work if you use it:
George, A, Kim, DN, Moser, T, Gildea, IT, Evans, JE, Cheung, MS. Graph identification of proteins in tomograms (GRIP-Tomo). Protein Science. 2023; 32( 1):e4538. https://doi.org/10.1002/pro.4538
See the included license and disclaimer files
August George, PNNL, 2022