openmm_workflow

Here you will find a conda environment which forms a workflow for running common molecular dynamics tasks in openmm. To install perform the following:

1. Clone this repository and cd to it

   git clone [email protected]:miro-astore/openmm_workflow

   cd openmm_workflow

• (optional) if you are on a HPC or you use modules it is recomended you make sure you load CUDA such as using module load cuda

2. Install dependencies and this environment using conda.

   conda env create -f environment.yml

To perform an unbiased molecular dynamics in a slurm environment do the following.

1. Copy the necessary .inp files.

cp openmm_workflow/min_relax.inp .

cp openmm_workflow/unbiased_production.inp .

• (optional) edit these files to reflect your desired minimisation, equilibration and unbiased molecular dynamics protocols.

2. Copy the necessary .slurm files if you are in a slurm environment. It is recquired that you edit these scripts to reflect your hpc and python environments. The scripts included here only serve as examples.

cp openmm_workflow/min_relax_equil.slurm .

cp openmm_workflow/unbiased_production.slurm .

2. Submit the minimisation and relaxation script which will start the workflow.

sbatch min_relax_equil.slurm

Openmm is extremely flexible, further customisation to the molecular dynamics protocols used in this package can be added by copying the base scripts like unbiased.py and umbrella.py in the src directory of this package.
They can then be run in the following way after modification.

python customised_openmm_script.py -ff forcefield -i custom_inputs.inp 

To add further inputs to the molecular dynamics environment through the .inp interface, any extra parameters need to be reflected by a change in src/openmm_workflow/utilities/omm_readinputs.py.