RE: readytowear SILVA 138 Training Error

[imonteroo]

readytowear SILVA 138 Training Error

that's correct! the reference sequences, taxonomy, and weights all need to have matching IDs. the plugin RESCRIPt has an action for filtering a taxonomy, so you can use that to remove IDs not in the sequences. you will need to re-create the animal distal gut weights after doing that (and there is a script in the readytowear repository showing how to do this). Alternatively, you could just train a full-length 16S classifier with those weights. A full-length classifier loses a little bit of ac…

Hi Nicholas. Could you please explain ahow to filtering a taxonomy, so you can remove IDs not in the sequences and re-create human-stool weights?

[SoilRotifer (Mike Robeson)]

Hi @imonteroo,

If you run qiime rescript filter-taxa --help, you'll see that you can simply provide the sequence file to the --m-ids-to-keep-file parameter. Then you provide your taxonomy file with the --i-taxonomy parameter. This will remove any taxonomy with IDs that are not found within the sequence file.

[imonteroo]

Thank you Mike. I tired the next code

qiime rescript filter-taxa --m-ids-to-keep-file silva-138.1-ssu-nr99-seqs-V3V4-uniq.qza --i-taxonomy readytowear-master/data/silva_138_1/full_length/human-stool.qza --o-filtered-taxonomy human-stool-V3V4.qza

but "human-stool.qza" is not a FeatureData[Taxonomy] file, is an FeatureTable[RelativeFrequency].

I am trying to create a silva-138.1-ssu-nr99-V3V4-stool-classifier.qza

I could create silva-138.1-ssu-nr99-V3V4-classifier.qza but I don't have human-stool-V3V4.qza so error "Number of priors must match number of classes" appears.

I could create silva-138.1-ssu-nr99-stool-classifier.qza because I have humann-stool.qza with full 16S from redytowear. I couldn't find the V3V4 version.

I am analyzing 16S V3V4 Illumina sequences so I think I need the V3V4 trained classifier.

[Nicholas_Bokulich]

Hi @imonteroo ,
If you are trying to create new weights with a different taxonomy you will need to follow this tutorial from the beginning:

Using q2-clawback to assemble taxonomic weights

Note: This guide assumes you have installed QIIME 2 using one of the procedures in the install documents. This tutorial extends the q2-feature-classifier tutorial to show how you can improve taxonomic classification accuracy by assembling ecologically informed taxonomic weights for your samples using q2-clawback. This approach is described in this article: https://doi.org/10.1038/s41467-019-12669-6 We will retrain the naive Bayes classifier from the q2-feature-classifier to classify the ski…

The readytowear weights are built using specific taxonomies (as described in the readytowear repository) so will only work with those taxonomies.

Good luck!