autoencoders_project

Unbiased simulations:

unbiased_simulations

Contains Python scripts for OpenMM simulations of:

• 2-D particles on a Muller-Brown potential energy surface
• Alanine dipeptide in water

Biased simulations

biased_simulations

Contains Python scripts for OpenMM simulations interleaved with rounds of CV discovery for 2-D particles on a Muller-Brown potential energy surface

Example simulations

example_simulations/

This directory contains Python scripts for OpenMM simulations of test 2-D and 3-D systems, demonstrating various capabilities of OpenMM.

Systems modelled:

• Protein in water
• Methane in water, configuration and topology generated using gmx pdb2gmx and gmx solvate utilities
• Box of water
• Box of gaseous Lennard-Jones particles
• 2-D Lennard-Jones particles

Test simulations with openmm-torch:

openmm_torch_test_simulations

Contains Python scripts for test simulations using OpenMM-torch to add biasing forces on

• 2-D particles on a Muller-Brown potential energy surface
• Alanine dipeptide in water
• Methane in water

Reference autoencoder model:

autoencoder

Contains code for PyTorch model to train an autoencoder on an MB particle trajectory

Modified GROMACS topology files (for OpenMM to read)

GMX_TOP

Can modify topology files in this directory and include them in OpenMM simulation scripts.