Skip to content

shirleyahn/we_analysis_scripts

Repository files navigation

we_analysis_scripts

Some of the scripts have been originally prepared by Terra Sztain, Anthony Bogetti, John Russo, and others from Prof. Lillian Chong's group (University of Pittsburgh) and have went through several modifications.

  • simtime_walltime.py: prints out the total simulation time and walltime for the WE simulation of interest using west.h5
  • w_pdist.sh: plots average free energy landscape and evolution of progress coordinate values - useful to keep track of progress of your currently running WE simulation (useful tip: periodically make copies of west.h5 in case the main one gets ruined)
  • autocorrelation_function.py: calculates the autocorrelation of the property of interest, such as arrivals from state A to state B (arrivals.txt). the confidence interval is generated using a Monte Carlo bootstrapping strategy where a boostrap consists of 1000 randomly drawn datasets (with replacement) (printed from script). this can be used, for instance, to see where the first point of the autocorrelation (autocorrelation_function.txt) falls within the confidence interval and can be used to calculate the correlation time.
  • contact_freq folder: includes scripts that extract salt bridges / hydrogen bonds that have lengths less than 3.5 angstroms (get_target_trajs.py and note that this particular salt bridge / hydrogen bond was kept track of as auxdata during the WE simulation) and plot contact frequency (probability distribution function) over one of the progress coordinates in WE (pdf.py).
  • durations folder: includes scripts that extract iteration #s and segment #s that correspond to when and which segment/walker made the transition of interest (e.g., going from A to B), the weights of those corresponding segments/walkers, and the duration of those transitions. The main script to execute is plot_duration.py. Make sure to change the TODOs in plot_duration.py. The script will also generate a distribution plot of durations in PDF form.
  • free_energy_plots folder: includes scripts that can plot smoother, more customizable free energy plots compared to the default WESTPA free energy plots
  • pdist_plots folder: includes scripts that generate bins for w_pdist when values are negative (default w_pdist will not work in this case so use bins.py). The user can copy and paste the output from bins.py as the first line for w_pdist_rbdc_elec.sh. Make sure that plotting.py is pulling the same data as in bins.py!
  • rate_plots folder: includes scripts that can extract rates from direct.h5 (from w_ipa -ao -W west.h5) and calculate the average from three independent runs
  • traj_concatenate folder: includes scripts that concatenate iterations to make a trajectory. Often times, we will end up with a list of iteration #s and segment #s (e.g., from durations folder) and will want to make trajectories of each transition. traj.sh will trace a particular iteration # and segment # so that it obtains all of the past conformations of this particular segment/walker. runtraj.sh is the main script to execute with a list of iteration #s and segment #s like up_pathways.txt.
  • w_crawl folder: includes scripts for w_crawl, which "crawls" through west.h5 to extract properties from trajectories, such as alpha carbon coordinates, etc. runcrawl.sh is the main script to execute/submit and wcrawl_functions.py is the main script to edit to obtain the property of interest. Afterward, copy_h5_dataset.py ran be used to concatenate the resulting h5 file to west.h5.

About

No description, website, or topics provided.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published