S-shooting in OpenPathSampling

Software Technical Information

The information in this section describes OpenPathSampling as a whole. Information specific to the additions in this module are in subsequent sections.

Language
Python (2.7)
Documentation Tool
Sphinx, numpydoc format (ReST)
Application Documentation
http://openpathsampling.org
Relevant Training Material
http://openpathsampling.org/latest/examples/
Licence
LGPL, v. 2.1 or later

Authors: Andreas Singraber

This module implements the S-shooting method [1] in OpenPathSampling.

Purpose of Module

S-shooting [1] is a recently developed method to determine rate constants of rare events. It is similar in spirit to the reactive flux method but its relaxed requirements help to overcome practical problems. The method is based on a simple shooting algorithm where trajectories are propagated forward and backward in time for a fixed number of timesteps. The starting points need to be provided and must lie in the saddle point region. This so-called S region (hence the name S-shooting) is defined via a suitable reaction coordinate and must to separate the stable states A and B in such a way that no trajectory can connect A with B without visiting S. In contrast to the reactive flux method the time derivative of the reaction coordinate is not required, which makes this approach applicable to systems exhibiting diffusive dynamics along the reaction coordinate. The S-shooting method can also be applied if the initial shooting points are taken from a biased simulation. Thus, it is a natural follow-up to free energy calculations like umbrella sampling and, in combination with free energy curves, allows the computation of rate constants.

The implementation of the S-shooting method in OpenPathSampling (OPS) is split into two main parts:

  • Forward and backward trajectories started from initial snapshots are harvested and glued together calling the SShootingSimulation class. The user needs to provide the initial snapshots, a suitable definition of the S region and the desired trajectory length.
  • The S-shooting analysis is performed upon calling the SShootingAnalysis class. Mandatory arguments include the definition of the stable states (A and B) and of the S region. In case the initial snapshots are taken from a biased simulation a bias function may be provided as an optional argument.

This module comes also with an IPython example notebook demonstrating the method by applying it to a one-dimensional system (a brownian walker in a double-well potential).

[1](1, 2) Menzl, G., Singraber, A. & Dellago, C. S-shooting: a Bennett–Chandler-like method for the computation of rate constants from committor trajectories. Faraday Discuss. 195, 345–364 (2017), https://doi.org/10.1039/C6FD00124F

Background Information

This module builds on OpenPathSampling, a Python package for path sampling simulations. To learn more about OpenPathSampling, you might be interested in reading:

Testing

Follow these steps to test the module:

  1. Download and install OpenPathSampling (see http://openpathsampling.org/latest/install.html).

    Caution

    This module has been developed alongside a specific OPS version available at that time. If incompatibilities arise as OPS is further enhanced, please use version 0.9.5 available here: https://github.com/openpathsampling/openpathsampling/releases/tag/v0.9.5 .

  2. Install the nose package.

  3. Download the source files of the module (see the Source Code section below).

  4. Install the module: change to the S-Shooting directory and run python setup.py install.

  5. Run the tests: execute nosetests in the S-Shooting directory.

Examples

See the sshooting-example.ipynb IPython notebook in the source directory, here is the direct link: https://gitlab.e-cam2020.eu/singraber/S-Shooting/blob/master/ops_s_shooting/sshooting-example.ipynb To run the example execute jupyter notebook sshooting-example.ipynb in your terminal.

Source Code

The source code for this module is located here: https://gitlab.e-cam2020.eu/singraber/S-Shooting

Tip

Ultimately, this module will be merged into the official OPS code. Check the status of the corresponding pull request here: https://github.com/openpathsampling/openpathsampling/pull/787 .