Error messages for dbBact

Community Plugin Support
1

Dear Sir or Madam:
I am trying to us the q2-dbBact plugin and with no success. The commond and plugin error is below:

qiime dbbact enrich-pipeline --i-table repseqsfiltered/table192.qza --m-metadata-file metadata/sample-metadata.tsv --p-field VacState --output-dir resultdbbact/enrichpipline
creating logger
Plugin error from dbbact:

Table seems to contain hashes and not sequences. Please supply the --i-repseqs representative sequences file, or create the table (deblur/dada2) with the --p-no-hashed-feature-ids flag.

Debug info has been saved to /tmp/qiime2-q2cli-err-aw_c1yqk.log

qiime dbbact enrich-pipeline --i-table repseqsfiltered/table192.qza --i-repseqs repseqsfiltered/repseqs.qza --m-metadata-file metadata/sample-metadata.tsv --p-field VacState --output-dir resultdbbact/enrichpipline
creating logger
Plugin error from dbbact:

No matching primers found. Are the reads reverse-complemented?

Debug info has been saved to /tmp/qiime2-q2cli-err-ezzfv22_.log

qiime dbbact enrich-pipeline --i-table repseqsfiltered/table192-uf.qza --i-repseqs repseqsfiltered/repseqs192-uf.qza --m-metadata-file metadata/sample-metadata.tsv --p-field VacStates --output-dir resultdbbact/enrichpipline
creating logger
Plugin error from dbbact:

No matching primers found. Are the reads reverse-complemented?

Debug info has been saved to /tmp/qiime2-q2cli-err-2bs07p5d.log

qiime dbbact heatmap --i-table repseqsfiltered/table192.qza --i-repseqs repseqsfiltered/repseqs.qza --i-taxonomy resultTaxonomy/taxonomy.qza --m-metadata-file metadata/sample-metadata.tsv --p-sort-field VacState --o-visualization resultdbbact/heatmap192
creating logger
Plugin error from dbbact:

object of type 'DNAFASTAFormat' has no len()

Debug info has been saved to /tmp/qiime2-q2cli-err-l2ho3eqx.log

qiime dbbact draw-wordcloud-vis --i-table repseqsfiltered/table192.qza --i-repseqs repseqsfiltered/repseqs.qza --o-visualization resultdbbact/wordcloun192
creating logger
Plugin error from dbbact:

object of type 'DNAFASTAFormat' has no len()

Debug info has been saved to /tmp/qiime2-q2cli-err-p472tf8y.log

qiime dbbact enrichment --i-diff resultdbbact/diffdsfdr.qza --p-source dsfdr --o-enriched resultdbbact/enricheddsfdr
creating logger
Plugin error from dbbact:

Input file contains sequence hashes instead of actual sequences.
Please use qiime dbbact enrichment-hash and supply the rep_seqs.qza.

Debug info has been saved to /tmp/qiime2-q2cli-err-nb0odw22.log

I am using the plugin within qiime2-2021.2.

Regards,

2

Hi Yang,
Thanks for your feedback. Here are suggestions about each error encountered:

  1. You probably created your feature table without the --p-no-hashed-feature-ids flag, and therefore it does not contain the actual ASV sequences but rather hashes. Since dbBact contains information about ASV sequences, it needs the actual sequences. Therefore you need to also provide the rep-seqs file (as you did in 2) with the --i-repseqs parameter.

  2. Ok, so now you provided the rep-seqs and it should work. But it seems your sequences maybe are not from a primer region supported in dbBact. dbBact currently supports the v1, v3, and v4 forward primers. A few questions:
    2a. Can you provide a few ASV sequences from the rep-seqs file (do: tar -xvf REP-SEQS-FILE.qza , and then go into the resulting directory/data and copy from the fasta file, or alternatively attach the rep-seqs file here)?
    2b. what version of q2-dbbact are you using (do: qiime q2-dbbact --version)
    2c. Also, can you re-run the enrich-pipeline command with the --verbose option and post the output?

  3. What is the difference between the table192 and table192-uf files? same as 2, can you provide a sample of the ASV sequences there?

  4. This was a q2-dbbact bug. Thanks for pointing it out!
    We have fixed it and uploaded a new version. Can you update your q2-dbbact version to v1.3.0 (using pip install --upgrade q2-dbbact) and rerun?
    Also, you can try running the heatmap command on the sample files:

And running the command:

qiime dbbact heatmap --i-table cfs-table.qza --i-repseqs cfs-rep-seqs.qza --i-taxonomy cfs-taxonomy.qza --m-metadata-file map.cfs.txt --p-sort-field Subject --o-visualization test-heatmap.qzv

Did it work?

  1. Same bug as 4 (now fixed), you can also try running the wordcloud on the sample files specified in 4:
    qiime dbbact draw-wordcloud-vis --i-table cfs-table.qza --i-repseqs cfs-rep-seqs.qza --o-visualization test-wordcloud.qzv

  2. Similar to 1, please specify the rep-seqs file (like you did in 2)

Let me know how it goes,
Amnon

3

Hello Amnon,
Thanks for the response and explanations.

I upgraded the plugin from version 1.1.1 to 1.3.0 and tried again. According to the error message of the new version, it is the sequence data that caused the errors.

The sequences were between primers 515F and 806R for 16S rRNA gene and several bases at the beginning were trimmed. I am wondering whether such sequences could be used for the enrich pipeline.

regards

4

Hi Yang,
unfortunately, dbBact only works with reads originating from one the forward primers V1, V3 or V4.
Why were bases at the beginning trimmed? How many bases were trimmed? I would recommend re-running the denoising step without the initial trimming at the beginning (from my experience, read quality is almost always good at the beginning). Then look and see if the resulting denoised ASVs look good (using the dbbact heatmap or other methods).

good luck, and let me know how it goes
Amnon

5

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