QIIME 2 2021.2 is now available

The QIIME 2 2021.2 release is now available! Thanks to everyone involved for their hard work!

As a reminder, our next planned QIIME 2 release is scheduled for April, 2021 (QIIME 2 2021.4), but please stay tuned for updates.

Check out the QIIME 2 2021.2 docs for details on installing the latest QIIME 2 release, as well as tutorials and other resources. Get in touch on the QIIME 2 Forum if you run into any issues!

Virtual machine builds will be available sometime within the next week - watch this topic thread for an update!

BREAKING CHANGES

q2-feature-classifier
- classify-sklearn: the default value for the reads_per_batch parameter was changed from 0 to "auto" and no longer allows 0. If you specified the reads_per_batch parameter explicitly as 0, you will need to change it to "auto" instead of 0. If you were not explicitly setting this parameter then you have nothing to worry about!

Please update scripts, workflows, etc. as appropriate. Stuck? Reach out here on the forum for help.

Here are the highlights of the release:

QIIME 2 Framework
- @thermokarst made some changes related to the Usage API, which is still a work-in-progress, more on this soon!
  - refactored the DiagnosticUsage driver to take a more idiomatic approach -
  - added a missing API method (comment) to the base driver
  - injected the action helpers into the base driver
docs
- @andrewsanchez & @jwdebelius fixed a typo or two in the PD Mice tutorial
- @andrewsanchez updated our documentation repo to accommodate new requirements introduced by a Sphinx update.
q2cli
- @thermokarst refactored this interface's Usage API driver to take a more idiomatic approach - the Usage API is still a work-in-progress, more on this soon!
q2-types
- @misialq refactored the DNAFASTAFormat into a more general FASTAFormat that should streamline the addition of new sequence formats (think e.g. RNA) and derive their Aligned versions
- @misialq added a new ProteinFASTAFormat and a corresponding ProteinSequence type to enable protein analysis - first plugin(s) using those are in the works
q2-feature-table
- @dwt & @thermokarst Added filter-opposite-axis as an option to filter-samples and filter-features. If filter-opposite-axis is False features/samples on the opposite axis which have zero frequency after the filtering will not be removed.
- @BenKaehler, @thermokarst, and @andrewsanchez added an average overlap_method to merge and enabled merge to process RelativeFrequency FeatureTable s.
q2-quality-control
- updated transformers to use view api to transform FeatureData[Sequence] to pandas.Series
q2-gneiss
- @mortonjt added ilr_phylogenetic_ordination to enable unsupervised analysis on a phylogenetic tree using the ILR transform
- @mortonjt added ilr_phylogenetic_differential to allow for Differentials produced from differential abundance tools such as Songbird and Aldex2 to converted to phylogenetic coordinates through the ILR transform.
- @mortonjt added tutorials on how to use these commands can be found on the empress github repo
busywork
- @thermokarst automated incrementing copyright years in our various QIIME 2 · GitHub repositories.
- @thermokarst migrated QIIME 2 · GitHub repositories off of TravisCI and onto GitHub Actions
workshops
- @thermokarst fixed a bug that prevented us from creating new workshop entries on the site.
- @thermokarst changed the sorting logic on the homepage to allow long-running workshops to be listed as upcoming, rather than "past"
q2-feature-classifier
- @Oddant1 made an improvement to the classify-sklearn action so that it is more clear when "auto" reads_per_batch ing is engaged.
  - * See the breaking changes section for more details.
q2-diversity
- @collin-le contributed a nice metadata validation enhancement so that a user-friendly error is presented when NaNs are present in the metadata.