reweight_residue_secstruct

Reweighting of GaMD (Amber)-sampled residue secondary structure propensities of peptides

Please see the following publication for more information on the applied discrete residue-based secondary structure propensity reweighting approach:

"Simulation of the Positive Inotropic Peptide S100A1ct in Aqueous Environment by Gaussian Accelerated Molecular Dynamics"; M. Glaser, N. J. Bruce, S. B. Han, R. C. Wade, J Phys Chem B. 2021 May 13;125(18):4654-4666 (https://pubs.acs.org/doi/10.1021/acs.jpcb.1c00902)

The reweighting procedure is implemented in the Python module src/reweight_residue_secstruct/reweight_residue_secstruct.py (please also check the introductory comment in reweight_residue_secstruct.py and its options for more information).

Python libraries needed for reweight_residue_secstruct.py

To run reweight_residue_secstruct.py, please first create the necessary Conda environment reweight_residue_secstruct via the supplied YAML file (environment_reweight_residue_secstruct.yml)

$ conda env create -f environment_reweight_residue_secstruct.yml

Example case on how to use reweight_residue_secstruct.py

The example case provided in the directory example allows:

1. to create the input files for reweight_residue_secstruct.py from Amber GaMD output files
2. to run reweight_residue_secstruct.py with the generated input files

The example directory contains:

• Amber GaMD production run trajectory files (gaMD_params_stripped.nc, gaMD_prod_1_stripped_cut.nc)
• the corresponding Amber GaMD log files (gamd.log, gaMD_prod_1_cut.log), containing the GaMD boost potential information for the respective MD steps the frames were outputted at
• two bash scripts: prepare_input_reweight_residue_secstruct.bsh and run_reweight_residue_secstruct.bsh

Caution

The example Amber GaMD output files were generated with Amber18. Please note that the example bash scripts as well as reweight_residue_secstruct.py were only used in combination with GaMD output from Amber18.

Important

To rerun the example, just remove the directories input_reweight_residue_secstruct and output_reweight_residue_secstruct. Also make sure you have a working Amber installation, as cpptraj is needed. The example bash scripts were setup and tested in the context of the above mentioned reweight_residue_secstruct Conda environment as well as cpptraj from Amber18.

To run the example, do the following:

1) Setup your Amber installation, if not already done:

$ source /your-path/amber18/amber.sh 

2) Activate the Conda environment reweight_residue_secstruct:

$ conda activate reweight_residue_secstruct

3) Next, execute the following two bash scripts within the example directory:

$ ./prepare_input_reweight_residue_secstruct.bsh
$ ./run_reweight_residue_secstruct.bsh

About prepare_input_reweight_residue_secstruct.bsh:

This bash script will generate the necessary input files for reweight_residue_secstruct.py, dssp.out and weights.dat, from the GaMD trajectory and GaMD log files, respectively. The bash script will write its output to the directory input_reweight_residue_secstruct. The file dssp.out contains the secondary structure information for each peptide residue of all the GaMD frames, obtained via cpptraj. The file weights.dat contains the corresponding boost potential information parsed from the GaMD log files in the format that is also used by the standard GaMD reweighting scripts PyReweighting-*D.py, which are used for reweighting of continuous order parameters.

Note

Please consult the comments in prepare_input_reweight_residue_secstruct.bsh for more information on its input and the generated output etc.

In the example case, the script prepare_input_reweight_residue_secstruct.bsh aggregates the information from a consecutive GaMD production run (i.e., a single replicon), split over two trajectory files. In case of the first trajectory file (i.e., gaMD_params_stripped.nc), only the last third of frames is taken (as the remaining part corresponds to preparatory steps).

Important

• If your trajectory is split over more than two trajectory files, you have to extend the script accordingly, in any case, make sure that the order of your GaMD frames and their respective GaMD log file information are in sync.
• A dual boost potential was applied during GaMD simulations (relevant for parsing of GaMD boost potential information to generate the file weights.dat).

Tip

By making the corresponding adaptions to prepare_input_reweight_residue_secstruct.bsh, in principle, one can also aggregate information from different GaMD replica for reweighting.

About run_reweight_residue_secstruct.bsh:

The bash script is a wrapper-script to run reweight_residue_secstruct.py. It will collect all the output from reweight_residue_secstruct.py in the directory output_reweight_residue_secstruct.

About reweight_residue_secstruct.py:

Note

• Please check comments in run_reweight_residue_secstruct.bsh for information on the reweight_residue_secstruct.py output files.
• The script reweight_residue_secstruct.py checks for common standard Amber force fields amino acid three-letter codes, if your residue name is not listed there, the allowed_residues-list needs to be extended accordingly.

Tip

The script reweight_residue_secstruct.py can be run in parallel by using its -m parallel option. With this option, depending on the available cores, as many residue secondary structure reweightings as possible are then performed simultaneously.