Classifier does not support confidence values

[Merlin (Merlin Szymanski)]

Hello all,

i have problem with the feature-classifier, showing the error code
“Classifier does not support confidence values”

–> I used reference reads and reference taxonomy from the Silva 132 release (99% Alignment, full taxonomy)
–> I extracted reads, but didnt trim (since sequences are variable in length):
–> I created a classifier from the extracted reads (fit-classifier-naive-bayes)
–> I tried to assign taxonomy to my sequences

I did run the --verbose command. This is what I got:

(qiime2-2018.6) qiime2@qiime2core2018-6:~/Desktop/shared/Classyfier$ qiime feature-classifier classify-sklearn --i-classifier 02_classifierJulia.qza --i-reads Importierte\ Rohdaten/SequenzenJulia.qza --o-classification Classfication --verbose

Traceback (most recent call last):
File “/home/qiime2/miniconda/envs/qiime2-2018.6/lib/python3.5/site-packages/q2cli/commands.py”, line 274, in call
results = action(**arguments)
File “”, line 2, in classify_sklearn
File “/home/qiime2/miniconda/envs/qiime2-2018.6/lib/python3.5/site-packages/qiime2/sdk/action.py”, line 232, in bound_callable
output_types, provenance)
File “/home/qiime2/miniconda/envs/qiime2-2018.6/lib/python3.5/site-packages/qiime2/sdk/action.py”, line 367, in callable_executor
output_views = self._callable(**view_args)
File “/home/qiime2/miniconda/envs/qiime2-2018.6/lib/python3.5/site-packages/q2_feature_classifier/classifier.py”, line 212, in classify_sklearn
reads, classifier, read_orientation=read_orientation)
File “/home/qiime2/miniconda/envs/qiime2-2018.6/lib/python3.5/site-packages/q2_feature_classifier/classifier.py”, line 169, in _autodetect_orientation
result = list(zip(*predict(first_n_reads, classifier, confidence=0.)))
File “/home/qiime2/miniconda/envs/qiime2-2018.6/lib/python3.5/site-packages/q2_feature_classifier/_skl.py”, line 45, in predict
for chunk in _chunks(reads, chunk_size)) for m in c)
File “/home/qiime2/miniconda/envs/qiime2-2018.6/lib/python3.5/site-packages/sklearn/externals/joblib/parallel.py”, line 779, in call
while self.dispatch_one_batch(iterator):
File “/home/qiime2/miniconda/envs/qiime2-2018.6/lib/python3.5/site-packages/sklearn/externals/joblib/parallel.py”, line 625, in dispatch_one_batch
self._dispatch(tasks)
File “/home/qiime2/miniconda/envs/qiime2-2018.6/lib/python3.5/site-packages/sklearn/externals/joblib/parallel.py”, line 588, in _dispatch
job = self._backend.apply_async(batch, callback=cb)
File “/home/qiime2/miniconda/envs/qiime2-2018.6/lib/python3.5/site-packages/sklearn/externals/joblib/_parallel_backends.py”, line 111, in apply_async
result = ImmediateResult(func)
File “/home/qiime2/miniconda/envs/qiime2-2018.6/lib/python3.5/site-packages/sklearn/externals/joblib/_parallel_backends.py”, line 332, in init
self.results = batch()
File “/home/qiime2/miniconda/envs/qiime2-2018.6/lib/python3.5/site-packages/sklearn/externals/joblib/parallel.py”, line 131, in call
return [func(*args, **kwargs) for func, args, kwargs in self.items]
File “/home/qiime2/miniconda/envs/qiime2-2018.6/lib/python3.5/site-packages/sklearn/externals/joblib/parallel.py”, line 131, in
return [func(*args, **kwargs) for func, args, kwargs in self.items]
File “/home/qiime2/miniconda/envs/qiime2-2018.6/lib/python3.5/site-packages/q2_feature_classifier/_skl.py”, line 52, in _predict_chunk
return _predict_chunk_with_conf(pipeline, separator, confidence, chunk)
File “/home/qiime2/miniconda/envs/qiime2-2018.6/lib/python3.5/site-packages/q2_feature_classifier/_skl.py”, line 68, in _predict_chunk_with_conf
raise ValueError(‘this classifier does not support confidence values’)
ValueError: this classifier does not support confidence values

Plugin error from feature-classifier:

this classifier does not support confidence values

Thanks a lot,

Merlin

[Mehrbod_Estaki]

Hi @Merlin,
Hmm, not really sure what to make of this error message yet but a couple of things first:
Can you confirm that SequenzenJulia.qza is the representative-sequences from your denoising/clustering outputs and not your actual feature-table? The former being the correct file.
Could you also share with us your classifier artifact (02_classifierJulia.qza) please.
Thanks!

[Merlin (Merlin Szymanski)]

Hello Mehrbold,

thank you for your answer!
SequenzenJulia.qza is the FeatureData[Sequences] file (unimported a FASTA file with ca. 40 sequences having no taxonomic assignment).

Then the classifier: 02_classifierJulia.qza (20.9 KB)

I solved the "confidence value" problem, however, by disabeling confidence values. Still other problems occured, so I believe there is a problem with my classifier. I guess I will try to use another SILVA database or another setting for training the classifier. Dont think too much about it, since I am using Qiime just as an additional tool for my bachelor thesis, it is not so tragic, if it wont work.

Greetings,

Merlin

[Mehrbod_Estaki]

Hi @Merlin,

I’m still not sure what or where your SequenzenJulia.qza file is coming from. But this file needs to be the reprsentative-sequences you obtained from the OTUclustering/denoising methods. In the q2 tutorials this would be the rep-seq.qza. This is the file you use along your classifier to assign taxonomy and NOT your actual feature table.
40 sequences in your rep-seqs seems pretty low, unless this is a mock community or something with very low diversity?
The steps you’ve taken for training your classifier seems right to me though double check to make sure those are indeed your primers, the forward primer seems a bit too short and on a quick google search I couldn’t find those matches. I could just not be looking thoroughly though, worth a double check.

Also, you can use the pre-trained SILVA classifier from the data-resource page, see if the issue is indeed coming from your classifier.
If you’d like help with the other problems you’re running into just give us a bit more detail and we’ll try our best to sort them out.
Good luck!