Filtering FeatureSequence and FeatureData to retain specified data

[baehsung (HEE-SUNG BAE)]

Hi all,

I have gotten the "rep-seqs" and "table" files the from DADA2 denoising step.
Below table is the rep-seqs data visualized.

would anyone mind giving me a comment how to remove some specific seqs (for example, the seqs selected in the picture) from those files primarily obtained from dada2 steps?

I am reading the tutorial on filtering data (Filtering data — QIIME 2 2022.2.0 documentation); however, i have not figure out how to make commands for this.

Best regards,

Hee-Sung

[colinbrislawn (Colin J Brislawn)]

Hello Hee-Sung,

I think you could remove those by their IDs using identifier based filtering.

Is there something specific these have in common, because there may be a better way to select and remove all of them based on a common trait.

[baehsung (HEE-SUNG BAE)]

Thanks Colin,

I have tried to use the "identifier based filtering"; however, i met several errors from several trials. I am still struggling with this issue.

Meantime, i am trying to select seqs in specific size range, i.e., 300-400 bp, by eliminating <300 bp and >400 bp of seqs from the feature seqs and table. Do you have an idea about this?

Best regards,

HS

[colinbrislawn (Colin J Brislawn)]

Good morning,

If you would like to post your errors, we can take a look and see what's happening.

@baehsung

i am trying to select seqs in specific size range, i.e., 300-400 bp, by eliminating <300 bp and >400 bp of seqs from the feature seqs and table. Do you have an idea about this?

When I'm trying to control the length of my amplicons, I usually perform this trimming or filtering at the very start of the process after importing into Qiime2. You could set these thresholds using dada2 before denoising, instead of modifying your ASV table. That's what I would suggest trying next!

[baehsung (HEE-SUNG BAE)]

thanks Colin,

I agree with you on primarily setting all parameters like the length of seqs before denoising process.

HS