da-barplot feature request

Hello,

I have installed the amplicon-2025.10 release and looked into the q2-composition plugin to see whether a interface change in the da-barplot methods made it to the new release :slight_smile:

From coming up next in 2025.4:

  • The ancombc2-visualizer action will be removed, in favor of da-barplot.
  • da-barplot will be updated to produce the visualization now produced by ancombc2-visualizer, and it will be updated to accept results of ancombc and ancombc2 as inputs.

I tested the 2025.10 da-barplot with my data and got this warning:

/home/NAME/miniconda3/envs/qiime2-amplicon-2025.10/lib/python3.10/site-packages/unifrac/init.py:9: UserWarning: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html . The pkg_resources package is slated for removal as early as 2025-11-30. Refrain from using this package or pin to Setuptools<81.
import pkg_resources

I assume that this is unrelated to da-barplot since the output visualization is created and it is in the old layout. I assume that a da-barplot update has not made into the 2025.10 release.

This gives me the chance to ask for a minor improvement in an updated da-barplot plugin/method. Since the ancombc2-visualizer features will go into the update (as stated above), may I suggest to expand the subset features options in a way that lfc (log fold changes) can be filtered in both directions at the same time (e.g. lfc greater than 2 OR lfc less than -2) either be allowing an OR operator or by allowing the absolute value (lfc greater than |2|).

Best regards,

4 Likes

Hello @arwqiime,

Sorry for the delay on this. We have an issue for the absolute value request here, apologies that that wasn't possible already. We'll plan to get it done some time during this release.

The renaming of ancombc2-visualizer to da-barplot has become somewhat deprioritized due to the effort required to make ancombc2-visualizer compatible with the older ancombc output formats, but we are still planning to get around to it eventually.

Thanks,
Colin

4 Likes

Right now, having to run the visualizer twice (once for positive, once for negative) or manually filtering the feature table beforehand adds unnecessary friction when you just want a quick overview of the strongest drivers.

2 Likes

@jakabaesj10, we agree - we're working on a fix for this.

1 Like