The git repository for this content is hosted at https://github.com/SCALE-MS/workshop
Workshop material includes content originally presented through https://github.com/kassonlab/gmxapi-tutorials, derived from the example in Figure 1 of 10.1371/journal.pcbi.1009835
Sample input files for these examples have been shared from previous research projects. They are covered by separate copyright and licensing details.
BRER workflow sample inputs: DOI 10.5281/zenodo.5122931
FS peptide (input_files/fs-peptide/
):
Sorin and Pande, Biophys J. 2005 Apr; 88(4): 2472–2493; doi:10.1529/biophysj.104.051938 (used with permission).
Get the gmxapi 2022 software stack
pip install gmxapi
after installing and activating GROMACS 2022Refer to installation instructions at the respective project websites.
Note that scalems
depends on RADICAL Pilot.
If you use pip
to install scalems
, radical.pilot
and its dependencies will be installed automatically.
Refer to https://scale-ms.readthedocs.io/en/latest/install.html for more information.
If you prefer to use a containerized installation of the software and examples,
and you are comfortable with Docker,
see the docker directory in this repository
or pull scalems/example-complete
from
DockerHub.
First, install GROMACS 2022, create a Python virtual environment, and install the gmxapi
Python package (see above))
git clone --recurse-submodules https://github.com/SCALE-MS/workshop.git
cd workshop
. /path/to/venv/bin/activate
. /path/to/gromacs/bin/GMXRC
pip install gmxapi
pip install -r external/scale-ms/requirements-testing.txt
pip install external/scale-ms
pip install -r requirements.txt
pip install .
examples/
If you have Docker installed,
you can build an image from the included Dockerfile
or docker pull scalems/example-complete
.
See docker/README.md for more information.
Warning:
The --rm
in the command line above tells Docker to remove the container when you shut down the container.
Any edits you make to the notebook will be lost.
If you want to save your changes,
the best choice is to use the File->Download option from within the notebook.
Alternatively, you could explicitly make a snapshot of the container with
docker commit
.
You could run docker
without the --rm
option,
but you will need to clean up extra containers manually to keep from filling up your hard disk.
Workflow is not checkpointed. You are advised to use a clean working directory for each script invocation.