Greengenes2 taxonomy-from-table error

Library Plugin Support
1

Hello!

I'm using the new greengenes2 plugin with qiime2-2023.2 to process v4 microbiome data. Data have gone through demux and denoise processes. I used the following code and came across this error:

qiime greengenes2 filter-features
--i-feature-table table.qza
--i-reference 2022.10.taxonomy.asv.nwk.qza
--o-filtered-feature-table table.filtered.qza

Saved FeatureTable[Frequency] to: table.filtered.qza

qiime greengenes2 taxonomy-from-table
--i-reference-taxonomy 2022.10.taxonomy.asv.nwk.qza
--i-table table.filtered.qza
--o-classification gg2.taxonomy.qza

Plugin error from greengenes2:

No requested tips found

Debug info has been saved to /var/folders/9n/q0zly08x30l1bfz2vybrjwd40000gn/T/qiime2-q2cli-err-_ofz3oc4.log

I also tried a few other methods and got the same error:

qiime greengenes2 taxonomy-from-features
--i-reference-taxonomy 2022.10.taxonomy.asv.nwk.qza
--i-reads rep-seqs.qza
--o-classification gg2.taxonomy.qza

Plugin error from greengenes2:

No requested tips found

Debug info has been saved to /var/folders/9n/q0zly08x30l1bfz2vybrjwd40000gn/T/qiime2-q2cli-err-_euwvm4l.log

wget http://ftp.microbio.me/greengenes_release/2022.10/2022.10.taxonomy.md5.nwk.qza

qiime greengenes2 filter-features
--i-feature-table table.qza
--i-reference 2022.10.taxonomy.md5.nwk.qza
--o-filtered-feature-table table.filtered2.qza

Saved FeatureTable[Frequency] to: table.filtered2.qza

qiime greengenes2 taxonomy-from-table
--i-reference-taxonomy 2022.10.taxonomy.md5.nwk.qza
--i-table table.filtered2.qza
--o-classification gg2.taxonomy.qza

Plugin error from greengenes2:

No requested tips found

Debug info has been saved to /var/folders/9n/q0zly08x30l1bfz2vybrjwd40000gn/T/qiime2-q2cli-err-6r_1w9xc.log

I would appreciate any recommendations!

8

Hi @cannon.320,

Can you describe the upstream denoising parameters used, as well as the specific 16S primers?

On an aside, I've opened an issue oriented at improving the information provided from filter-features when matches don't occur.

All the best,
Daniel

9

Hi Daniel,

Thanks for your help. 505F and 806R primers. Here's my code from the denoising:

qiime dada2 denoise-paired
--i-demultiplexed-seqs tai-demux.qza
--p-trim-left-f 5
--p-trim-left-r 5
--p-trunc-len-f 247
--p-trunc-len-r 231
--o-table table.qza
--o-representative-sequences rep-seqs.qza
--o-denoising-stats denoising-stats.qza

qiime feature-table summarize
--i-table table.qza
--o-visualization table.qzv
--m-sample-metadata-file tai.mapping.txt

10

Thanks, @cannon.320!

So in this case, it is unlikely the ASVs are directly represented in Greengenes2. The primary set of ASVs placed are 90, 100, and 150nt in length off of 515F as that is the most common processing type in Qiita across 300,000 public/private 16S samples. We've also only placed the forward read.

Just to check, is it 505F or 515F?

There are a few options:

  • If it's 515F, you could try not performing a trim on 5', and truncating 3' so the ASVs are within one of the lengths we typically place.

  • 16S ASVs regardless of region can be classified with classify-sklearn using the full length model, if within 515F-806R against that region specifically.

  • 16S ASVs can also be clustered against Greengenes2 using the non-v4-16s action, which in turn executes q2-vsearch's cluster-features-closed-reference against the Greengenes2 backbone.

Let me know how it goes!

All the best,
Daniel

1 Like
11

Success! I used the non-v4-16s directions and it worked. Thank you so much for your advice, I greatly appreciate it!

(And yes, it was 515F, apologies for the typo.)

2 Likes
13

Great!!! Glad it worked :slight_smile: