Quality control and denoising

Hello to everyone,

I am a new QIIME2 user. I have some problems during quality control and denoising procedure and I would appreciate it if anyone can help me out, please.

I try to explain everything but if there is anything need to be added, I would be happy to do that.

I have 64 samples (more than 10 million reads, Forward and Reveres) and they are demultiplexed.

Based on the Interactive Quality plot that I have received I am trim and trunc as following but the results is not satisfying. I was wondering maybe I do not have enough overlapping. If it is the case what can I do? (And how to know If I have enough overlapping?)
I write down the commands that I was running for quality control and denoising using dada2:

qiime dada2 denoise-paired
--i-demultiplexed-seqs demux-paired-end.qza
--p-trim-left-f 0
--p-trim-left-r 0 \
--p-trunc-len-f 210 \
--p-trunc-len-r 184
--o-table table.qza
--o-representative-sequences rep_seqs.qza
--o-denoising-stats denoising_stats.qza

Hi @ari_sh70,

Welcome to the forum!

Your reads are relatively low quality, especially your revers reads. Your trim position looks good for the denoising, you’re keeping a fiar number with trimming and denoising. You could even possibly relax the parameters a bit - you would lose more reads during denoising but could maybe keep some during merging. But, ultimately, I think you’re going to find the quality of your reverse reads is just too low to be able to retain and join them. So, one suggestion is to try and keep as much of your forward reads as you can (maybe trimming the first 20 and then around 250 or 260) and denoise there. It’s recommended but not required that you use paired end reads. And, a lot of people have published on forward reads alone.


1 Like

Dear Justine,

Thank you for your information. I will follow based on what you have said and hope I could get a better result.

Thank you,


1 Like

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