ommtk 0.2.3

OpenMM Toolkit by Redesign Science

OMMTK

OpenMM Wrappers for General and Enhanced MDSimulations

This repo contains several simulation classes wrapped around openMM infrastructure that allow for rapid devlopment and usage of various MD Simulations.

Currently Supported are the following simulations: Minimization, Standard MD (base class); Heating and Equil MD simulations (NVT and NPT respectively), Gentle Heat Sims (heat/release restraints over a range of stages), simulated tempering, metadynamics and metadynamics with focused sampling on ligand unbinding (Segment Walker Sims). The simulations support a full API with a range of functions and easily adapted run statements for specialized protocols.

Installation

Dependencies:

OMMTK requires very few dependencies. Below is a list you can install with conda or pip:

python 3.6 or higher
simtk.openmm
numpy
mdtraj

To install the latest stable version we recommend using pip with the following command (this is the only method guaranteed to provide a stable version).

pip install ommtk

Documentation

The most recent documentation: https://redesignscience.github.io/OMMTK/

Quickstart run

    # Load packages
    import parmed
    from ommtk import MDSimulation, select_atoms, unit
    
    # Note: Requires parameterized parmed in memory named parmed_structure!

    #instantiate sim from parmed alone, minimize (save h5 file) and run
    mdsim = MDSimulation(parmed_structure=parmed_structure)
    mdsim.minimize()
    minimized_parmed = mdsim.run(2 * unit.nanosecond)

    #build sim using a previous trajectory or h5 file
    mdsim = MDSimulation(parmed_structure=parmed_structure, coordinates=trajectory.h5)

    #build sim with positional restraints, run some equilibration and remove them for longer run
    protein_atoms = select_atoms(parmed_structure=parmed_structure, keyword_selection='protein')
    mdsim = MDSimulation(parmed_structure=parmed_structure, atoms_to_restrain=protein_atoms, restraint_weight=4)
    mdsim.run(1*unit.nanosecond)
    mdsim.update_restraint_weight(2)
    mdsim.run(1*unit.nanosecond)
    mdsim.remove_positional_restraints()
    end_parmed = mdsim.run(4 * unit.nanosecond)