Deblur hashed feature IDs

Hello,
I’m running the commands (sown at the end) and realized that in the visualization outputs of

  • qiime metadata tabulate
  • qiime deblur visualize-stats
    contain the full feature-ids which I used to load the reads;

However the

  • qiime deblur visualize-stats
    output visualization contains only truncated feature ids; I have used the “–p-no-hashed-feature-ids” parameter, but this does not solve the problem.

01_U341F-U806R-NW3-1
10_U341F-U806R-NW3-11
11_U341F-U806R-NW3-12
(…)
get truncated to
01
10
11
(…)

qiime deblur denoise-16S
–i-demultiplexed-seqs 2a_Biogas3b-all_deblur-filtered.qza
–p-trim-length 272
–o-representative-sequences 2b_Biogas3b-all_deblur-rep-seqs.qza
–o-table 2b_Biogas3b-all_deblur-table.qza
–p-sample-stats
–p-jobs-to-start 12
–o-stats 2b_Biogas3b-all_deblur-stats.qza

qiime metadata tabulate
–m-input-file 2a_Biogas3b-all_deblur-filter-stats.qza
–o-visualization 2c_Biogas3b-all_deblur-filter-stats.qzv # full feature ids present

qiime deblur visualize-stats
–i-deblur-stats 2b_Biogas3b-all_deblur-stats.qza
–o-visualization 2d_Biogas3b-all_deblur-stats.qzv # full feature ids present

qiime feature-table summarize
–i-table 2b_Biogas3b-all_deblur-table.qza
–o-visualization 2e_Biogas3b-all_deblur-joined-table.qzv
–m-sample-metadata-file metadata-Biogas3b.txt

This command gives me an error (The following IDs are not present in the metadata: ‘01’, ‘02’, ‘03’, ‘04’, ‘05’, ‘06’, ‘07’, ‘08’, ‘09’, ‘10’, ‘11’, ‘12’, ‘13’, ‘14’, ‘15’), which is true since the metadate file contains the full feature ids.
If I run it without “–m-sample-metadata-file metadata-Biogas3b.txt”, the command completes but the visualization has now the truncated feature ids.

So, I assume that the previous artifacts di still have the full feature ids.
Does anybody have an idea what could cause the shortening of the feature ids?
Best regards

Hey there @arwqiime!

Those are sample IDs, not feature IDs, which is why the no-hashed-feature-ids parameter made no difference.

What are the sample IDs in your demux summarize viz? How did you import? If you used a fastq manifest format can you please share your CSV?

Hi,

the import was:
qiime tools import
--input-path NGS655b/manifest-Biogas3b.csv
--output-path Biogas3b_demux.qza
--type SampleData[PairedEndSequencesWithQuality]
--source-format PairedEndFastqManifestPhred33

I have attached the manifest file, and the demux summarize viz.

manifest-Biogas3b.csv (2.1 KB)
Biogas3b_demux.qzv (286.5 KB)

Thank you for your support!

Hey there @arwqiime. This appears to be an issue in deblur — underscores in sample IDs are not supported. You can reimport your samples using dashes (or another character) in place of the underscore. We have an open issue for this in q2-deblur, the plugin wrapping deblur. Let us know what you wind up doing. :qiime2: :t_rex:

Hi @thermokarst, I will modify the manifest file an reimport the sequences.
Thank you for you support!

This solved my problem! Underscores should not be in the manifest file…

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