Second Opinion on trimming length


I attached my quality plot. The quality of reverse reads was not too good. I am confused on deciding the trimming length.

As for now, Ive used 280 bp(forward) and 170 (reverse). Will that be ok?

Hey there @steffi — to be honest, it is pretty difficult to interpret these data through a photo of your monitor — maybe you could attach the QZV, instead? Your numbers look okay, but it would be helpful to be able to see the 7-number summaries interactively… :eyes:

2 Likes

I have attached my qzv format. demux.qzv (286.4 KB)

Ive executed with above mentioned trimming length. I looked into the file table.qzv table.qzv (322.2 KB).
I found that sequence counts per samples are very low.
I have 15 samples and counts vary from 35,087 to 25.
Is it because of my trimming length? I used untrimmed reads for dada2 analysis.
Based on the quality plot, I hope increasing in trimming length for reverse reads will increase the noise. What should i do? Should I proceed with the forward reads? or is there any way I can rectify this issue?

Thanks for the QZV, @steffi!

From the forward reads, I would think about dropping the first ~7 nts and everything after ~228 (approx Q30 median). From the reverse reads, I would look at dropping the first ~13 nts and everything after ~164 (approx Q30 median).

My guess is "yes" --- you left a lot of lower-quality read in, which is less-than-ideal for DADA2.

Try my recommendation above (also, I noticed you ran with n-threads set to 1) --- if you can afford upping that number I would set it to 4 if you can --- DADA2 will run much more quickly for you.

Keep us posted! :qiime2: :t_rex:

1 Like

Oops, @Mehrbod_Estaki just pointed out to me that those values are probably too extreme, and will likely result in no overlap. Let’s see what you get out the other side and go from there… :t_rex:

1 Like

Exactly. Because my amplicon size is more than 500. May I go with my forward reads

1 Like

Sure - keep us posted!

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