bgmol

----- WORK IN PROGRESS ------

A collection of OpenMM systems and data generated for those systems.

This package is an extension of the openmmtools.testsystems module with a twofold aim:

It contains a collection of OpenMM systems (with initial positions and topologies).
It provides infrastructure for sharing data (positions, forces, energies, velocities) generated for these systems.

Quickstart: Install

Make sure that you are in a conda environment that has OpenMM and mdtraj installed.

Clone the code

git clone git@github.com:noegroup/bgmol.git

And install

cd bgmol
python setup.py install

Quickstart: Example

from bgmol.datasets import Ala2Implicit300
dataset = Ala2Implicit300(download=True, read=True)

The dataset contains forces, energies and coordinates it also holds a reference to the system that defines the potential energy function.

openmmsystem = dataset.system

The system is an OpenMMSystem object, it provides access to the simtk.openmm.system instance, the topology, and a set of initial coordinates. For example, we can run an OpenMM simulation as follows

from simtk.openmm.app import Simulation, LangevinIntegrator
integrator = LangevinIntegrator(dataset.temperature, 1, 0.001)
simulation = Simulation(openmmsystem.topology, openmmsystem.system, integrator)
simulation.context.setPositions(openmmsystem.positions)
simulation.step(10)

The dataset contains coordinates (xyz), forces, and energies.

dataset.energies

Adding Systems to the Repository

The preferred way to add systems is by adding a subclass of OpenMMSystem to the bgmol.systems module, see examples therein.

Adding Samples to the Repository

TBD

Copyright

Acknowledgements

Project based on the Computational Molecular Science Python Cookiecutter version 1.2.

drorlab/bgmol