Thanks for your help! I figured out a few things so wanted to share here in case it helps anyone out in the future:
One thing I was having difficulty with was that using the new Greengenes2 plugin was providing feature IDs for my ASVs that were inconsistent with the feature IDs from the DADA2 denoising table - so I had no way to match up the tax ID with the FASTA sequences from the rep-seqs.qza file. I stumbled across another thread of someone who had the same issue (see here: ASV table and Taxonomy table have different formats with greengenes2 - #2 by wasade) and it looks like it may be an incompatibility issue when trying to use original QIIME2 features with the new Greengenes2 plugin. When I just used the QIIME2 classifier to assign taxonomy the feature IDs were consistent and I was able to link the rep-seqs.qza to my tax IDs from Greengenes2.
E.g., replaced Greengenes2 plugin code:
qiime greengenes2 non-v4-16s
--i-table denoise_table_filtered.qza
--i-sequences denoise_seqs_filtered.qza
--i-backbone 2022.10.backbone.full-length.fna.qza
--o-mapped-table gg2_table.qza
--o-representatives gg2_seqs.qza
qiime greengenes2 taxonomy-from-table
--i-reference-taxonomy 2022.10.taxonomy.asv.nwk.qza
--i-table gg2_table.qza
--o-classification gg2_table_taxonomy.qza
with QIIME2 classification code:
qiime feature-classifier classify-sklearn
--i-reads denoise_seqs_filtered.qza
--i-classifier gg_2022_10_backbone.v4.nb.qza
--o-classification gg2_table_taxonomy_sklearn.qza
qiime tools export
--input-path gg2_table_taxonomy_sklearn.qza
--output-path gg2_taxa
Something interesting I noticed from doing this was how different my resulting number of classified features are between the 2 above methods - and I don't really understand why.
With the Greengenes2 plugin I ended up with only 602 classified ASVs in the gg2_table.qza from 8,282 input ASVs from DADA2.
With the QIIME2 sklearn classification method I ended up with almost all of my ASVs classified (8,237/8,282).
Is there a reason this could be happening? Am I totally missing something?
Thanks again for your help 