Dada2 denoise-single for reverse reads

I had 2 runs of 16s amplicon sequencing. The new run is using 341F for R1 and 806R for R2, but our old run was using 806R for R1 and 515F for R2, which had caused additional problems. My goal to merge these two runs for all downstream analysis using “feature-table merge” and “feature-table merge-seqs”. In order to do so, it looks like my only choice is treat it as single end 806R reads. I am planning to using R2 from new run and R1 from old run to run “dada2 denoise-single” process separately, but I have noticed by default “dada2 denoise-single” is using forward reads, no option for choosing reverse reads. Can someone share insight?
Even so, due to low quality issue of new run at R2 806R, I probably can only use --p-trunc-len 180. Any suggestion on this?

16s_New_demux.qzv (287.9 KB)
16s_Old_demux.qzv (285.9 KB)

Hi @hotblast,

For ASV methods (or any method that isn’t strictly reference based), I would advise against merging studies that used different primer sets, because there are typically biases (i.e. certain taxa will be picked up better by one primer set than another one).

If you still want to continue with the merge (at your own risk!), you could potentially use the forward reads for both, and trim the 341F reads to match the same start as the 515F reads. Combining just reverse reads can be more problematic because they usually have much lower quality scores than forward reads (as you noted).

1 Like

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