Container-based installation of
PyMol, with interaction through the browser via
ipymol and Jupyter notebook (based on
jupyter/notebook). A convenient and portable way to render pretty pictures of molecules, and much more.
The installation also contains
scipy among other things, so it can be used for a variety of scientific Python tasks.
First, note down the IP address of the currently running Docker machine (which is called
dev in this example) with
docker-machine ip dev
Download the image from Docker hub :
docker pull ocramz/jupyter-docker-pymol
You can see the list of locally available Docker images with the command
Run the image :
docker run --rm -it -p 8888:8888 ocramz/jupyter-docker-pymol
Point your browser to the IP address of the Docker machine found initially, and port 8888, i.e.
<docker-machine-ip>usually starts with
A Jupyter session should appear in the browser at this point. Up and running !
- Within Jupyter, start a Python 3 document (or just start by modifying the provided example notebook)
Declare inline figure rendering within Jupyter notebooks and setup the connection to PyMol:
%pylab inline from ipymol import viewer as pm pm.start()
Run your PyMol tasks, e.g. :
pm.do('fetch 3odu; as cartoon; bg white;') f1 = pm.show()
Docker (Windows and OSX users should install the Docker Toolbox : https://docs.docker.com/toolbox/overview/)
docker-machinerunning in the current shell (setup guide : https://docs.docker.com/machine/get-started/)
This project uses PyMol 22.214.171.124 and Python 3
At present, this setup is intended for local use only (i.e. the Docker image, along with all the computational payload i.e. PyMol and the Python interpreter, is running on the same host that runs the browser).
There is NO authentication to the notebooks and the Jupyter user is
the PyMol project contributors , pymol.org
the Jupyter and iPython project contributors , jupyter.org
Carlos Hernandez for
ipymol , https://github.com/cxhernandez/ipymol