This package provides various tools for modelling proteins with cubic symmetry in Rosetta and EvoDOCK. It can be used as a library but also contains import scripts for 2 purposes:
- Generating cubic symmetry files for use in Rosetta and EvoDOCK
- Generating full biological assemblies from the output of Rosetta and EvoDOCK
NOTE: This package is installed alongside EvoDOCK!
For a standalone install the following must be installed:
- Python-3.6 or later (PyRosetta dependency).
- PyRosetta (http://www.pyrosetta.org) (Can be installed with Anaconda). You need to obtain a license before use (see the link)
- MAFFT (https://mafft.cbrc.jp/alignment/software/) (can be installed with Anaconda/brew/apt)
Clone the cubicsym repository and cd into it. Then run the install script.
git clone https://github.com/Andre-lab/cubicsym.git
cd ./cubicsym
pip setup.py install You should be all set.
scripts/cubic_to_rosetta.py generates cubic symmetry files of either Icosahedral, Octahedral or Tetrahedral symmetry.
The usage of the script is well documented. Use python cubic_to_rosetta.py --help to see more.
There are 2 ways to run the script. An automatic way and a manual way.
When using the automatic way one only needs to specify --structures and --symmetry (see options below).
A test can be run (inside the cubicsym dir) with
python scripts/cubic_to_rosetta.py --structures tests/inputs/1stm.cif --symmetry I --symdef_outpath tests/outputs/ --input_outpath tests/outputs/ --rosetta_repr 1 --rosetta_repr_outpath tests/outputs/ --overwriteIn case the automatic way fails, one can use the manual way. When using the manual way one needs to specify all or some of the following:
- --hf1
- --hf2
- --hf3
- --f3
- --f21
- --f22
'hf' stands for highest fold which corresponds to the highest symmetrical fold for the system which for an icosahedral structure is the 5-fold, for the octahedral structure the 4-fold and for the tetrahedral structure the 3-fold. 'f3' stands for the 3-fold (which all cubic structures have) and f21 and f22 the two 2-folds. Subunit numbers are parsed to each of these options to determine the symmetry as they are assigned to by the script. To see the subunit numbers, first run the script with the flag: --output_generated_structure. This will generate an output file of the full biological assembly. Look at the output in a structural program (like PyMOL or Chimera) and from it assign the subunit numbers to the options. These numbers should always be related to the main subunit.
A test to create an output structure to assign subunit numbers from can be run with:
python scripts/cubic_to_rosetta.py --structures tests/inputs/7m2v.cif --symmetry I --output_generated_structure --output_generated_structure_outpath tests/outputs --overwrite This will generate the following structure: tests/outputs/7m2v_generated.cif. From looking at that we can assign
the subunit numbers and run another test (inside the cubicsym dir) with the subunit numbers specified.
In most cases it is just enough to supply the 3-fold as this is what the script can struggle to predict sometimes:
python scripts/cubic_to_rosetta.py --structures tests/inputs/7m2v.cif --symmetry I --overwrite --f3 1 28 55 --symdef_outpath tests/outputs/ --input_outpath tests/outputs/ --rosetta_repr 1 --rosetta_repr_outpath tests/outputs/A test for the full description would look like this:
python scripts/cubic_to_rosetta.py --structures tests/inputs/7m2v.cif --symmetry I --overwrite --hf1 1 4 7 31 10 --hf2 55 52 58 16 49 --hf3 40 46 13 28 37 --f3 1 28 55 --f21 1 52 --f22 1 37 --symdef_outpath tests/outputs/ --input_outpath tests/outputs/ --rosetta_repr 1 --rosetta_repr_outpath tests/outputs/scripts/cubic_from_rosetta.py generates a full biological assembly files from the output of Rosetta or EvoDOCK.
The usage of the script is well documented. Use python cubic_form_rosetta.py --help to see more.
A basic test can be run with
python scripts/cubic_from_rosetta.py --structures tests/outputs/1stm_rosetta.pdb --symmetry_files tests/outputs/1stm.symm -o tests/outputs/ --overwrite