Help with PICRUSt Analysis Using Nanopore Metabarcoding Data

[Ana Benavides]

Hello everyone,

I am working with metabarcoding data obtained through Nanopore sequencing and need assistance with conducting functional prediction analysis using PICRUSt. Here are the details of my situation and what I have done so far:

Project Details:

• I have sequenced 16S rRNA samples using a Nanopore platform.
• I have completed taxonomic assignment using Kraken2.

Steps I have taken:

1. I have the FASTQ files and the taxonomic assignment results from Kraken2.

I need help with:

1. How to convert and prepare my FASTQ files to be compatible with QIIME 2 to obtain ASVs.
2. How to export the necessary files from QIIME 2 in a format compatible with PICRUSt.
3. The exact steps to run PICRUSt with my data, considering it originates from a Nanopore platform.
4. Any advice on specific adjustments I should make to ensure the accuracy of the results.

[Robyn Wright]

Hi there,

Apologies for the slow reply. 

I don't have experience with Nanopore data personally, so I don't want to give advice that's wrong, but it seems like this tool may be developed for working with Nanopore data and QIIME2.

For running PICRUSt2, you just need a fasta file containing the representative sequences (ASVs/OTUs) and a feature table containing the abundance of each of those representative sequences in your samples. You can see some example files within the tutorial. Aside from this, while as I said, I do not have any personal experience working with Nanopore data, I do not see any reason why you shouldn't then be able to run PICRUSt2 as normal - in fact, with the longer sequences, you should have a better chance of being able to place them accurately within the tree used. 

Robyn