OpenMM Continuity Force

This is a plugin for OpenMM that implements a custom many-particle force called a "ContinuityForce". This force aims to ensure that all particles in a set remain continuously connected in space. More precisely, if we imagine the particles as nodes of a network and draw an edge between particles that are closer than a cutoff distance ($d$), then the nodes would all be in the same component of the network. In other words, there exists a path of some length between any two nodes $i$ and $j$.

ContinuityForce achieves this by classifying particles into components on the fly and adding an attractive force between pairs of particles belonging to different components. The particles are chosen each dynamics step as the closest possible pairs between the two components. The attractive force is of the form $E(r)=k(r-d)^2$, where r is the separation between the particles and $d$ is a user-defined cutoff distance.

This was originally designed to aid FlexibleTopology simulations (paper). This plugin was made using the "OpenMMExamplePlugin" template.

Preparing the Environment

Use a CONDA environment for easy installation. Before installing this plugin, you should create this environment and install openmm:

conda install -c conda-forge openmm=8.1 cudatoolkit=11.8

Note: explicitly installing the cudatoolkit is now necessary for newer openmm versions. At this point you can also check if openmm can communicate with your GPU:

python -m openmm.testInstallation

where you should see that CUDA successfully computes forces.

You will also need swig:

conda install swig

And if you need to install cmake this can also be done using CONDA:

conda install cmake

Building/Installing The Plugin

To build and install openmmcontinuityforce, follow these steps:

1. Clone this repository to your own machine.

2. Create a directory in which to build the plugin, e.g. openmmcontinuityforce/build.

3. Change to the build directory, and run cmake: cmake .. and open ccmake: ccmake ..

4. Set OPENMM_DIR to point to the directory where OpenMM is installed. This is needed to locate the OpenMM header files and libraries. If you are using CONDA, this directory will be something like: /your/path/to/anaconda/envs/env_name.

5. Set CMAKE_INSTALL_PREFIX to the same directory.

6. If you plan to build the CUDA platform, make sure that CUDA_TOOLKIT_ROOT_DIR is set correctly and that BUILD_CUDA_LIB is selected.

7. Press "Configure" again if necessary, then press "Generate".

8. Type make install, followed by make PythonInstall. Note: there will likely be many warnings for PythonInstall related to OpenMMSwigHeaders.i. These can be safely ignored and as long as the build reaches 100%.

Test Cases

To run all the test cases build the "test" target by typing make test.

Accessing ContinuityForce in Python

You can use the plugin in your Python scripts:

from openmmcontinuityforce import ContinuityForce
force = ContinuityForce()

For an illustrative example, check out LJ_test.py in the examples folder.