Set of python scripts to extend the capabilities of the LIPMIX software for analyzing small-angle scattering data from lipid systems (as part of the ATSAS package), for example, at cluster systems with SLURM queue management system.
The scripts are provided "AS IS" and need to be adapted for the particular case, especially the initial parameters of the distributions and electron density peaks. Please refer to the manual of LIPMIX on the official webpage of the ATSAS package: https://www.embl-hamburg.de/biosaxs/manuals/lipmix.html.
If you use results from LIPMIX in your publication, please cite:
The overall idea is to try to fit a given scattering curve from unilamellar or multilamellar vesicles with different combinations of a number of bilayers. The main script writes the command file for the LIPMIX with the given combination of a number of bilayers expected and executes it in the assumption that the other parameters like positions and width of the electron density maxima do not change much. See the original paper and manual for a detailed description. For the small number of bilayers, this can be done. Note that the actual physical meaning of the result needs to be assessed separately.
python lipmix_multilayer_autofinder.py -l 1 -d 1 -i test_lipmix_MLV.dat
This command will create the simplest fit with one distribution and one bilayer. Data is shown for multilayer vesicles; therefore, the fit will not be optimal. Only for testing purposes.
Example curves are taken from SASBDB database.
python lipmix_multilayer_autofinder.py -l 3 -d 1 -i *.dat
This command will go over the list of *.dat files in the current directory and attempt to fit the data with a maximum of 3 layers and 1 distribution. Results will be saved in a separate directory with the name of the current *.dat file. This task can be run on the local computer and be executed in a reasonable amount of time. Useful for time series or temperature scans.
To obtain the best fits, one can run the following command in the current directory:
python lipmix_extract_min_chi_from_batch.py
Min chi: 1.7809 SASDG72_n_dist_1_[1_2_3]_7
Min chi: 6.9491 SASDG82_n_dist_1_[1_2_3]_7
The results can be found in the "result" directory.
python lipmix_multilayer_autofinder_SLURM.py -l 10 -d 1 -i test_lipmix_MLV.dat
This example tries to fit the multilayer vesicles data with up to 10 bilayers in different combinations. Parallel execution is recommended. Please check the options in the "Cluster submission" section of the code for the particular cluster in use. The example is written based on the SLURM submission system. This example is used only for demonstration purposes and is most suited for the small number of bilayers expected (~3-5).
A total number of permutations for ten bilayers can be calculated as follows:
python calc_number_of_permutations.py -l 10 -d 1
Total number of permutations: 1023
Other examples:
python calc_number_of_permutations.py -l 3 -d 1
Total number of permutations: 7
python calc_number_of_permutations.py -l 5 -d 1
Total number of permutations: 31
python calc_number_of_permutations.py -l 7 -d 1
Total number of permutations: 127
python calc_number_of_permutations.py -l 20 -d 1
Total number of permutations: 1048575
python calc_number_of_permutations.py -l 25 -d 1
Total number of permutations: 33554431
python calc_number_of_permutations.py -l 30 -d 1
Total number of permutations: 1073741823
The cluster batch example repeats the following for the local execution for demonstration:
python lipmix_multilayer_autofinder_SLURM.py -l 3 -d 1 -i *.dat
Since the SLURM submissions are happening one after another, one needs to check the available resources. In the case of ten files and ten expected bilayers, this can result in 10230 simultaneous jobs on the cluster.