Plugin error from taxa: Feature IDs found in the table are missing from the taxonomy:

before we used classify-consensus-vsearch, which has an option output_no_hits (default true) explicitly warning for differences between tax and featuer tables

Now we use classify-hybrid-vsearch-sklearn and we get Feature IDs found in the table are missing from the taxonomy

Is there possibly a difference in behaviour with respect to output_no_hits.

my quess: classify-hybrid-vsearch-sklearn does not output unclassified features. Is this correct? If so, how should I filter the feature table against the taxonomy?

I'm trying to build a bar plot after clustering and taxonomy assignment. I know there are older topics about the subject and there the issue was caused by filtering, However I am not filtering anything, but I'm using the classify-hybrid-vsearch-sklearn which is apparently in alpha stage (qiime 2025.4).

I get the following error error message

Plugin error from taxa:

Feature IDs found in the table are missing from the taxonomy: {'KJ578676.1.2010'}

These are my commands to get there:

clustering:

# DB_SEQ contains a fasta files with reference sequences

qiime vsearch cluster-features-open-reference \
        --i-table "run_20261010.ASV.table.qza" \
        --i-sequences "run20261010.rep-seqs.qza" \
        --i-reference-sequences "$DB_SEQ" \
        --p-perc-identity 0.97 \
        --p-threads 8 \
        --output-dir ./OTU

classification

qiime feature-classifier classify-hybrid-vsearch-sklearn \
                --i-query ./OTU/clustered_sequences.qza \
                --i-reference-reads "$DB_SEQ" \
                --i-reference-taxonomy "$DB_TAX" \
                --i-classifier "$DB_CLASSIFIER" \
                --p-perc-identity $CLASSIFIER_PERC_ID \
                --p-query-cov $CLASSIFIER_QUERY_COV \
                --p-threads 8 \
                --o-classification "./OTU/run20261010.taxonomy.qza"

barplot: where the error message appears

qiime taxa barplot \
        --i-table ./OTU/clustered_table.qza \
        --i-taxonomy ./OTU/run20261010.taxonomy.qza \
        --o-visualization ./OTU/run20261010.taxonomy.barplot.qzv

Hi @fenny ,

Please see the action description for the most likely explanation and some important limitations of this action:

  NOTE: THIS PIPELINE IS AN ALPHA RELEASE. Please report bugs to
  https://forum.qiime2.org! Assign taxonomy to query sequences using hybrid
  classifier. First performs rough positive filter to remove artifact and low-
  coverage sequences (use "prefilter" parameter to toggle this step on or
  off). Second, performs VSEARCH exact match between query and reference_reads
  to find exact matches, followed by least common ancestor consensus taxonomy
  assignment from among maxaccepts top hits, min_consensus of which share that
  taxonomic assignment. Query sequences without an exact match are then
  classified with a pre-trained sklearn taxonomy classifier to predict the
  most likely taxonomic lineage.

By default prefilter is true, so filtered features will be excluded from the classification step and missing in the final taxonomy. The solution is to filter your table to match the taxonomy e.g., with qiime feature-table filter-features.

Good luck!

Thanks, however, we are actually interested in low coverage, unclassified sequences. So maybe not the best classifier to use then