Dockerfiles for Computational Chemistry

This is a repository of Dockerfiles for software commonly used in computational chemistry:

• GROMACS (2020.4)
• ORCA (4.2.1, 5.0.1, 5.0.2)
• SIESTA (4.1.0)

Basically, this allows one to deploy a lightweighted, containerized environment in a matter of minutes, in which a particular package is ready to be used. Images are always based on the latest Ubuntu image available on Docker Hub. (How do I get started with Docker?)

How to run an image

All images are available in GitHub Container Registries, so the command below does everything for you:

$ docker run -it ghcr.io/geem-lab/IMAGENAME:TAGNAME /bin/bash

As a direct example:

$ docker run -it ghcr.io/geem-lab/gromacs:2020.4 /bin/bash

(Of course, you'll need Docker.) This will give you a minimal, fully functioning terminal with all required software available in /opt.

Note: if you build images locally (see below), you probably want to remove the ghcr.io/geem-lab/ part (see below).

How to build images locally

$ docker build -t IMAGENAME:TAGNAME IMAGENAME:TAGNAME

As a direct example:

$ docker build -t gromacs:2020.4 gromacs:2020.4

How to run images locally

$ docker run -it IMAGENAME:TAGNAME /bin/bash

As a direct example:

$ docker run -it gromacs:2020.4 /bin/bash

Useful technical references

• Install Docker Engine on Ubuntu
• Best practices for writing Dockerfiles
• Dockerfile reference