An integrated tool to score model PDB structures based on crosslink and monolink data.
Scoring for Crosslinks is performed using MNXL and Monolinks are scored by MoDS. If both crosslinks and monolinks are being scored the combination score XLMO is calculated.
Information on these scores can be found in Sinnott et al., Combining Information from Crosslinks and Monolinks in the Modeling of Protein Structures, Sturcture (2020), https://doi.org/10/1016/j.str.2020.05.012.
Run the following command to install the XLMTools and Jwalk modules:
python setup.py install
The setup.py script will install the relevant modules and set up 3 aliases in your ~/.bashrc:
DEPTH - The DEPTH algorithm for calculating residue depth, this in an external software from the Computational and Structural Biology Lab at Pune. Please note this program may not work on all OS, see http://cospi.iiserpune.ac.in/depth/htdocs/download.html for details.
jwalk - To the Jwalk.v2.1.py script for running Jwalk
xlm - To the xlmtools.v1.0.py script for running XLM-Tools
Flags can be listed by running xlm -h, which will show the below help message:
XLM Tools tools: A tool to score model protein structures according to crosslink and monolink data.
optional arguments: -h, --help show this help message and exit
XLM Arguments: -xl_list XL_LIST input list of crosslinks and monolinks
-jwalk_files [JWALK_FILES [JWALK_FILES ...]] jwalk files for MNXL scoring
-depth_files DEPTH_FILES [DEPTH_FILES ...] depth files for MoDS scoring
-outfile_name OUTFILE_NAME specify output file name
-sep SEP separator in output file, default is tab
-pdb PDB [PDB ...] specify pdb files for Jwalk/Depth run
Jwalk Arguments:
-jwalk flag to use if starting from .pdb files and running Jwalk
-vox VOX specify voxel size of grid
-surface use higher accuracy method to calculate solvent accessibility - requires Freesasa installation
-ncpus NCPUS specify number of cpus to use
Depth Arguments:
-depth flag to use if starting from .pdb files and running Jwalk
-depth_source DEPTH_SOURCE specify depth source
The only required argument is -xl_list, this contains the crosslinks and monolinks which the model structures are scored against. The file is structured as below:
aa1|c1|aa2|c2 <- Crosslink
aa1|c1|aa2|c2
aa1|c1|aa2|c2
aa1|c1|aa2|c2
aa1|c1 <- Monolink
aa1|c1
aa1|c1
Python 3.6
Scipy
Numpy
Biopython
- Matthew James Sinnott
- Dr Joshua Matthew Allen Bullock
- Prof. Konstantinos Thalassinos
- Prof. Maya Topf
This project is licensed under the GNU GPL v3.0 License - see the LICENSE file for details
If this software is used please cite:
Sinnott et al., Combining Information from Crosslinks and Monolinks in the Modeling of Protein Structures, Sturcture (2020), https://doi.org/10/1016/j.str.2020.05.012
- Dr David Holdershaw
- Dr Sony Malhotra
- Dr Mark Williams