Reproducibility of scientific data analysis is critical, but can be particularly challenging with computational analysis using cutting-edge open-source tools. Installation issues are often encountered, preventing end-users from accessing potentially useful tools for their analysis. One possible route to improving accessibility and reproducibility of analysis pipelines is the use of so-called “container” systems, such as Docker, Singularity and Apptainer. This introduction aims to demystify what containers are, why they’re useful, and how to use them. While using containers is not a silver bullet to solve all problems, it is a useful addition to a scientist’s toolkit that can assist in the sharing and use of robust analysis software. A talk in the CCP-volume EM "Show and Tell" series (https://www.ccp-volumeem.ac.uk/showandtell)
Martin Jones (Tue,) studied this question.