Deblur p-mean error value

Hello! Apologies in advance if this is kind of a dumb question but I'm having trouble finding information for a specific parameter of qiime2 deblur denoise-16s. I am following some older scripts that are not annotated or have any corresponding notes and the script uses deblur denoise code and specifies --p-mean-error 0.0017 (the full code is below). My question is why? Most of the example workflows and tutorials I've found online do not have this extra line in there and my google searches have been fruitless so far. They have another version of this script where the p-mean-error is 0.005, so I get that they're moving this threshold around but I don't understand it's purpose or why they're trying various values?

Any help is greatly appreciated! I'm working with paired-end 16s sequencing data!

qiime deblur denoise-16S
--i-demultiplexed-seqs step.01d.joined.filtered.qza
--p-trim-length 300
--p-mean-error 0.0017
--p-min-reads 10
--p-sample-stats
--o-representative-sequences step.02.v3.nomin.deblur.repseqs.qza
--o-table step.02.v3.nomin.deblur.table.qza
--o-stats step.02.v3.nomin.deblur.stats.qza

1 Like

Google has issues with our docs, for some reason, so it's not just you.

Here's the docs for deblur denoise-16S, including --p-mean-error
https://docs.qiime2.org/2023.2/plugins/available/deblur/denoise-16S/

You can read more about the deblur algorithm on GitHub. :open_book:

1 Like

Thank you so much for this, I couldn't find it on my own!

Follow up question: if I delete the --p-mean-error line completely does that mean it defaults with 0.005? Do I only need to enter this line if I'm changing the value to something other than 0.005?

That's correct.

You do need to pass [required] arguments, but the [defaults] are used unless told otherwise. The default for each setting is shown in the docs, and care is taken by the devs to choose reasonable defaults that should work well. (See this example of the devs discussing and documenting the default deblur values.)

1 Like