Very few remaining sequences after denoising some samples

[Agatsuma (Zenitsu Agatsuma)]

Thanks for your visiting!
I have two questions.
The first is I downloaded some SRR data from ncbi. The difference in the amount of data is relatively large. After converting to .fastq, denoising is performed. After denoising, the remaining sequence has good denoising effect, and there are few remaining sequences after partial denoising.I really dont know how to fix it.

The second question is about classifier.I would like to know how much of the sequence similarity will be classified into the same type during the annotation process and how to adjust the similarity threshold myself.
Thanks for your answer and discussion!

[lizgehret]

Hi @Agatsuma,

Thanks for reaching out! :qiime2:

Can you please provide the exact command you ran for your denoising procedure? This will help us to get a better idea of how many reads you are filtering out, and if this is something reasonable for your dataset.

With regards to your second question - since this is a slightly different topic (sample classification vs. denoising), can you please submit this as a separate post on the forum? This will help keep things organized for our moderators and forum users to better search for commonly asked questions.

Thanks!