Pac Bio amplicon clustering

[Toni]

Dear vsearch community,

I have a question regarding PacBio amplicon read processing using vsearch. Average length of the sequences in our project is ~700 bp but the  range is from 500 to 800 bp.

In another question from the forum (https://groups.google.com/g/vsearch-forum/c/qs_9Zo6cc8c) it was suggested that dereplication does not make sense for reads with variable length. Can I still use vsearch for quality filtering, chimera detection/removal and OTU clustering for my case or would you suggest some other tool to use? I know usearch/uchime/uparse was used in the past for similar project (same genes and sequencing technology) by other groups.

Also, since we have obtained hifi reads, quality scores are quite high (mean Q > 40) but would it make sense to perform any quality filtering prior to the rest of the processing?

I would really appreciate your advice!