Minimum information needed for publication

The data produced by qiime2 are the artifacts that contain all the information necessary to explain how the data were created. However, I’m afraid that the artifact files can hardly be added to publications, in particular to supplemetary data (if available) for several reasosn.
Does any of the q2 developers have an idea (or a suggestion/template), how a short, but sufficient explanation of the qiime2 analyses should be presented in the materials and methods section of publications?
Best regards


@arwqiime, a few thoughts on this.

That is definitely true for some journals, but note that the qza and qzv files are literally just zip files. If the journal accepts zip files as supplementary material, you could change the extension from .qza or .qzv to .zip, and submit the artifact as supplementary data.

If that’s not possible or you’re not interested in doing that, you can deposit your artifacts and visualizations in dryad or zenodo and get a DOI that can be shared in any journal, even in the main text (credit goes to @nbokulich for that suggestion).

Another approach would be to create a GitHub repository that contains your artifacts and any scripts that you created for that project, and include a reference to that in your paper. This doesn’t work very well if you have a lot of data though, but could be combined with deposition of data in dryad or zenodo (so scripts go in GitHub and data goes in dryad, for example).

You could also think about using the provenance graphs from your artifacts (viewable at to guide the design of an analysis workflow figure, where you illustrate and annotate all of the steps that were performed. This would be very powerful if, for example, the artifacts and visualizations at each step were annotated with the DOIs of the corresponding data stored in dryad.

Hope this helps to get you started thinking about this!

1 Like

@gregcaporaso, thank you for your suggestions. In particular, the possibility to present the provenance graphs seems great. I was not aware about this graph, because the embedded viewer of the virtual box package (2018.2) does not display the provenence graphs. Is it right that the provenence graph is only visible at

@arwqiime, currently those graphs are only visible at, but in the future you will have options for provenance viewing in other interfaces as well (in addition to lots of other cool provenance-related features).

1 Like

This topic was automatically closed 31 days after the last reply. New replies are no longer allowed.