pre quality-trimmed reads in qiime2.2022.11

[ireneferraris]

Hi,
I have two questions regarding 300 PE reads pre-trimmed by quality and then imported in qiime2.2022.11.

First, I was wondering if the different lengths among forward and reverse reads would be a problem to qiime2 in the merging phase. Moreover, I would like to know if their pre-trimming could affect diversity analysis eventhough they are paired end.

[llenzi (Luca)]

Hi @ireneferraris,

Welcome in the forum!

Can I ask you more details on your data? What so you meant with different length aming forward and reverse reads?

In general, any trimming of low-quality tails on forward and reverse read could affect the merging step, by reducing the overalpping region between the reads. Keep in mind that dada2 suggest at least 12 bp of overlapping. How much the trimming affect the merging step, it depends on the average length of the region you are working with, for example, if you dealing with 16S v4, r1 and r2 shoul doverlap a lot hence the low-quality trimming should not have agreat effect. If your region is v3-4, the trimming could have an impact.
You can see if you have enough overlap using the formula:
R1 length + R2 length - Amplicon length;
Eg, yout are trimming your sequence R1: 260; R2: 250; Amp length is 460 bp;

260 + 250 - 460 = 50

Your overlap is 50 bp on average, you are fine.

What you could do to see if the triming affects the diversity, is to analyse the sequences as paired-ends and using the forward read only. Any sequences that not merge is lost in the paired end analysis, while should be kept (as long as the quality is good enough) in the forward-only analysis.

Hope it helps,

Luca