florianvdent / qtools Goto Github PK

qtools is a toolset comprised of a Python library Qpyl and a set of command-line tools, for use with the molecular dynamics simulation package Q. This open-source project was created by Miha Purg (currently at Uppsala University) in order to provide Q-users with a set of well-tested tools for automating repetitive and/or tedious tasks, debugging simulations, and analysing results. The author strives to achieve a good balance between the level of user-control over Q (i.e., relatively low abstraction level) and the ease-of-use provided by qtools.

Disclaimer: qtools is not affiliated with Q nor should it be considered an official toolset for Q.

• Python API for interfacing with Q executables.
• EVB mapping and automatic fitting of off-diagonal and gas shift parameters.
• Support for quantum classical path and group exclusion calculations.
• Linear response approximation (LRA) calculations (including group contributions)
• Parameter conversions (OPLS, Amber/GAFF).
• FEP file generation for EVB simulations.
• Input generation for MD/FEP simulations and EVB calculations.
• Output parsing and analysis of MD, FEP and EVB.
• Graphical tool for visualizing results and debugging simulations:
  • Free energy profiles
  • LRA, REORG, group contributions
  • Energy breakdown (VDW, El, Bonds, ...)

The library and tools do not have any dependencies, except matplotlib which is required only for q_plot.py, a non-essential utility used mostly for troubleshooting simulations.

The only requirement is Python, version 2.7

Note: The command-line tools were designed to work in a typical high-perfomance-computer Unix environment, and have not been tested with Microsoft Windows.

Clone the repository to your local directory:

mkdir -p ~/bin && cd ~/bin
git clone https://github.com/mpurg/qtools

Add this line to your ~/.bashrc or ~/.bash_profile:

source $HOME/bin/qtools/qtools_init.sh

Run the CLI config script:

qscripts_config.py

The starting point should be the docs/tutorials folder. It contains working examples of how you can use the command line tools to prepare, simulate and analyse a simple SN2 reaction with EVB. Additionally, a recently published chapter in the serial Methods in Enzymology provides step-by-step instructions for performing EVB simulations on a real enzymatic system with qtools. Manuscript available here (paywalled).

For those inclined on using the Python API directly, jupyter-notebook examples calling the Qpyl API can be found in docs/examples. The API reference can be found in docs/api/html. Additionally, the tests, and the command-line interface (qscripts-cli) can be useful to explore the API.

Automated Testing is performed on Travis-CI, with pytest for Qpyl (code coverage). and simple regression tests for CLI tools. To run the tests locally, make sure you have pytest installed and simply type:

cd tests
./run_tests.sh

To acknowledge the use of qtools in scientific publications, please specify the version and the DOI of the latest release:

Example:
...analysis was performed with qtools v0.5.10 (DOI: 10.5281/zenodo.842003).

Miha Purg ([email protected])

Paul Bauer