About PICRUST2 database for training rep-seq

Hello,

I am writing this post to clarify some points while using the PICRUSt2 plugin in QIIME2. I am currently analyzing 16S rRNA gene data using a rep-seq trained naive Bayes classifier model based on the Greengenes2 database.

While reviewing the PICRUSt2 code, I noticed that the input files include the feature table and rep-seq files, but not the taxonomy.qza file used to train the rep-seq.

Therefore, I assume there must be a classifier within PICRUSt2 that has already been trained on some database. Could you please let me know which database it uses? Additionally, if the data is not based on Greengenes2, is there a way to use my taxonomy.qza file, which is trained on Greengenes2, with PICRUSt2?

Thanks.

Hi @bandy134,

I did a bit of poking around on the PICRUSt2 Github repository, but didn't find any information on which database was used (but this was really only a quick scan, so I may have missed it somewhere).

With that being said, I did find a note on the repo that suggests using this Google Group for all PICRUSt & PICRUSt2 questions - you may get a better response on there!

Best of luck :lizard:

thanks for your kind reply. Have a good weekend!!