How to get Greengenes reference into FeatureData[Taxonomy] format?

[Danielle_Wang]

Hi, I want to use the "Training feature classifiers with q2-feature-classifier" using the Reference Greengenes 13_8 99% OTUs from 515F/806R region of sequences

Here is the default code: but it doesn't work
qiime tools import
--type 'FeatureData[Taxonomy]'
--input-format HeaderlessTSVTaxonomyFormat
--input-path 85_otu_taxonomy.txt
--output-path ref-taxonomy.qza

How can I get the Greengenes Reference into FeatureData[Taxonomy] format?

Thanks.

[colinbrislawn (Colin J Brislawn)]

Hello Danielle,

Have you already discovered the Training feature classifiers tutorial? That includes a section on importing taxonomy txt files into the FeatureData[Taxonomy] format.

There is also a pipeline that automates lots of this process, which is also worth a look!

Let me know if you have any other questions!

[Danielle_Wang]

Hi, thanks for your reply!

I have tried the Training feature classifiers tutorial. That's where the problem comes in...

• my reference taxonomy is found here (4th one): Data resources — QIIME 2 2021.11.0 documentation.
• however, the format isn't compatible with the code that is documented in the tutorial. (it's not "HeaderlessTSVTaxonomyFormat")
• so the code gives me the error: (1/1) Invalid value for '--i-reference-taxonomy': Expected an artifact of at least type FeatureData[Taxonomy]. An artifact of type TaxonomicClassifier was provided.

When I directly tried to Train the Classifer with the Greengenes QZA file, apparently, the given Greengenes in an artifact of type TaxonomicClassifier. The expected format is FeatureData[Taxonomy].

I'm stuck. Do you know of a solution for this? Thank you so much.

[colinbrislawn (Colin J Brislawn)]

Oh, I get what's going on!

@Danielle_Wang

• my reference taxonomy is found here (4th one): Data resources — QIIME 2 2021.11.0 documentation.
  ...
  When I directly tried to Train the Classifer with the Greengenes QZA file, apparently, the given Greengenes in an artifact of type TaxonomicClassifier. The expected format is FeatureData[Taxonomy].

The data resources page provides pre-trained classifiers that have already been through this step. So you don't have to import data or train them, you can just use them directly on the classify-sklearn step.

That tutorial starts from step 1:

Because you downloaded a TaxonomicClassifier .qza file from the data resources page, you can skip to step 4.

Here's the full flowchart of the process, if that's helpful.