SAXS Z-score calculation
System requirements:
- Python 2.7
- SciPy (tested with v1.0.0, any recent version should do)
- NumPy (tested with v1.13.3, any recent version should do)
- IMP (optional, used to evaluate model-data fits, tested with v2.8.0)
Installation guide
Place the files calc_z.py and curves.npz to a folder (optionally within your system PATH) and call as a python script. No specific configuration is required.
The online server which requires no installation is available here.
Usage
python calc_z.py <data> [model]
Z0 score is calculated for an experimental scattering profile <data>. If a PDB file [model] is provided, the script will additionally output chi2 score of the fit and the corresponding Z-score.
Example
Example data can be found in the test folder. Run as "python calc_z.py original_0.dat original.pdb", it should take a couple of seconds on a modern PC. Expected outputs of the script:
Chi2 values
| SAXS profile | original.pdb | perturbed.pdb |
|---|---|---|
| original_0.dat | 0.90 | 1.14 |
| original_1.dat | 0.89 | 1.93 |
| original_2.dat | 0.98 | 4.18 |
| original_3.dat | 1.04 | 10.06 |
Z-scores
| SAXS profile | Z0 | original.pdb | perturbed.pdb |
|---|---|---|---|
| original_0.dat | 2.89 | 2.93 | 2.85 |
| original_1.dat | 3.34 | 3.37 | 3.14 |
| original_2.dat | 3.81 | 3.81 | 3.38 |
| original_3.dat | 4.15 | 4.14 | 3.47 |
The structural models and the SAXS profiles are based on the Tsirkone, V. G. et al. J. Biol. Chem. 292, 9699–9710 (2017).