Using lotus2 for Nanopore amplicon seq data

[Rahul Jain]

Hi Falk, 

I am working on some SSU amplicon data that I generated using Nanopore MinION platform. However, as of now I don't see ONT platform integrated in to the lotus2 pipeline. May I ask if it is possible to integrate this manually or use a nanopore specific SDM-config file. If yes, what should be the correct parameters to set in that config file? 

Many thanks already!

Rahul

[Falk Hildebrand]

Dear Raahul,

please see the github for lotus2 questions:  Issues · hildebra/lotus2 (github.com)  ( this forum is outdated and I only rarely check).

About Nanopore, this is not supported, because often you have ~95% seq accuracy and this is just difficult to deal with when you expect 97% OTU clusters. Quality filtering will likely remove most reads, and you wouldn't want that either. There are no current plans to integrate a dedicated Nanopore workflow, unless read qual is stably >Q20. However, if you want to try using LotuS2 nonetheless, I would recommend lowering the expected quality filters in the sdm config, allowing for more primer errors, deactivating accumulated errors and binomial estimates and using uparse for clustering.

Hth,

Falk