I've been using qiime for a while for research analyzing diet contents using eDNA (i.e., uncovering the taxonomy/species composition). Although I know qiime is meant for Illumina data, my data is nanopore sequences and I'm hoping others out there have used qiime with nanopore data too and can give me some guidance. The biggest issue I run into is that qiime filters sequences all to the same length but I need sequences to be different lengths. I have tried running vsearch outside of qiime to filter and cluster and then merged my data together and imported it back into qiime to run taxonomy analyses; However, I always run into issues, whether it be formatting or the code simply will just run forever without an output. Has anyone figured out a way to import multiple sequences of different lengths into qiime or have any suggestions for alternative pipelines/programs? I appreciate any feedback!
Helo Lauren,
Welcome to the forums! 
So, the good news is that Qiime2 supports many input data types, including multiple sequencing types. Like, the Qiime2 DADA2 plugin supports Illumina, 'pyro' sequencing 454 and IonTorrent, and PacBio's CCS sequencing.
https://docs.qiime2.org/2024.5/plugins/available/dada2/
However, the underlying problem remains:
Variable length amplicons are not well supported, including on Illumina!
Did you use PCR to target an amplicon before NanoPore sequencing or was untargeted metagenome sequencing performed?
Hi, thanks for getting back to me!
Yes, I used PCR to target an amplicon before Nanopore sequencing. More specifically, I ran PCR with 5 different primer sets to target specific diet contents (i.e., plant, animal, fish, etc.,), and then all PCR products for a single sample were combined and taken through library preparation steps before being sequenced on a Nanopore device.
Okay, wow! That sounds a lot like this Ion Torrent kit, which is also multi-region.
Multi-region kits have proven to be extraordinarily difficult.
Take a look at that thread and let us know what you want to try next.
There's another Lauren in that thread too, so maybe it's a curse! 
Great, thanks for your help! I'll check out that thread. I'm also going to try nanopores program and pipeline through epi2me that might work better for the varying sequence lengths, but we shall see!
Careful which EPI2ME workflow you use. Depends on your data type, but from what I can infer you want wf-metagenomics, not wf-amplicon.
This topic was automatically closed 31 days after the last reply. New replies are no longer allowed.