Targeted Gene Metagenomic Data Analysis    ◾    285

--o-table deblur/table_yoga_deblur.qza \

--o-stats deblur/stats_yoga_deblur.qza

The “--p-jobs-to-start” is an optional parameter and you can set it to the number of jobs to

run in parallel. To learn about more options, use “qiime deblur denoise-16S --help”.

The feature table and representative sequences artifacts will be used in the downstream

analysis.

The deblur stats summary artifact contains useful information about the filtering and

denoising. We can use the “deblur visualize-stats” plugin to generate a visualization file

(Figure 7.10).

qiime deblur visualize-stats \

--i-deblur-stats deblur/stats_yoga_deblur.qza \

--o-visualization deblur/stats_yoga_deblur.qzv

qiime tools view deblur/stats_yoga_deblur.qzv

Both clustering and denoising (with DADA2 or deblur) generate feature table and repre-

sentative sequences artifacts that can be used in the following analysis steps. You may need

to visualize these feature table and the sequence data artifacts. The “q2-feature-table” plu-

gin is used just for that. The “summarize” and “tabulate-seqs” methods are used to create a

visualization file for the feature table and representative sequences, respectively. As exam-

ples, the following script creates visualization files for the feature tables produced by the

de novo clustering and DADA2 denoising, respectively. The “summarize” method of the

“feature-table” plugin takes a feature table artifact and the sample metadata file as input

and it outputs a visualization file. We created the sample metadata file “sample-metadata.

tsv” earlier and we saved it in the “data” subdirectory.

FIGURE 7.10  Deblur denoising stats summary.