Requirement for fasta sequences and questions for when using Nanopore 16S data

[SandersonHaley]

I am working with Nanopore 16S metabarcoding data and did not produce FASTAs of the 16S sequences when getting the relative abundances, the pipeline went from read to an OTU table. There are about 45 different species in my samples. Could I just provide any complete 16S sequence from each of the species to the picrust2 (like from NCBI) or do I have to convert all my fastq/fast5 read files to fasta and provide that?

I also read on another forum that there are alignment issues with full 16S sequences when using picrust2. Is that still an issue? Is there a workaround for that now or another program I should be using instead?

Thanks, Haley S.

[gavinmdouglas]

Hey @SandersonHaley,

You could do that (although it was not tested with full-length 16S sequences in mind, so sometimes there can be alignment issues as you mentioned), but it's better to use representative sequences from your actual data, by which I mean final assembled 16S sequences.

Cheers,

Gavin