Data analysis from Oxford Nanopore

General Discussion
1

Hello,

I am analyzing Nanopore 16S data (24 samples), but my reads are much shorter than expected (300–500 bp instead of ~1500 bp).

What is the best QIIME2 approach in this case?

  • Should I use DADA2 or switch to OTU clustering (vsearch)?

  • How should I filter reads?

  • Can I still obtain reliable taxonomy (e.g., for Helicobacter pylori)?

Thanks!

2

Hello!
You should not use dada2 on Nanopore data. So you should use vsearch instead.
But you will end up with very high amount of unique OTUs due to the error rate of Nanopore data.

You can also try NaMeco, just adjust the minimum length to your data. You can import the output to Qiime2 after processing, together with taxonomy.

Best,
Timur

3 Likes