I am working with 2 groups of sequences. some of 300bp and the other of 150bp. How can I do them to work together?
i am using
qiime dada2 denoise-paired
Good afternoon @pufetin,
Are these all of the same amplicon from the same region? If so, you might be able to carefully choose trimming and truncating settings so that they all end up the same length after joining.
Maybe you could post the quality distribution of these reads after joining so we can check to see how close they are!
(You could also use qiime vsearch cluster-features-closed-reference to do closed ref clustering, which is not as good, but very easy.)
Let us know what you try! 
Colin
1 Like
thanks Colin
the problem is that if I truncate in 140pb those of 300, they don't overlap and it can't be joined
Can I work this step (denoise-paired) separately and join the output later?demux_seqs.qzv (292.0 KB)
1 Like
No. The truncation step is done for the purposes of denoising, and merging happens after denoising in that pipeline. You can adjust the truncation settings so that you do succeed in merging your reads.
Another alternative is to use q2-quality-filter to filter, q2-vsearch to merge the overlapping reads, then use q2-deblur to denoise (q2-dada2 will not work on pre-merged reads).
But this gets away from your original question. If @colinbrislawn's advice about trimming these to the same site does not work for you, then you should follow his advice about closed-reference OTU clustering as a very straightforward way to accomplish this. You could also check out q2-fragment-insertion for another method that allows you to compare these (specifically for beta diversity comparisons)
Good luck!
1 Like