Extract-reads no matches found

[Stephanieorch (Stephanie Orchanian)]

I’m unable to use the greengenes reference tree because I am analyzing the groEL gene rather than the 16S gene. We wanted a deeper analysis of the bifidobacterium genus and go to the species and even strain level that cannot be done using the 16S gene.

As recommended, I tried to use the naive Bayes classifier by running:

qiime feature-classifier extract-reads \
--i-sequences correct_no_spaces_only_bifido_and_gardnerella_cpn_rep_seqs_upper_no_replicates.qza \
--p-f-primer TCGTCGGCAGCGTCAGATGTGTATAAGAGACAGTCCGATTACGAYCGYGAGAAGCT \
--p-r-primer GTCTCGTGGGCTCGGAGATGTGTATAAGAGACAGCSGCYTCGGTSGTCAGGAACAG \
--p-trunc-len 490 \
--o-reads ref-seqs-cpn.qza

But I got this error:

Plugin error from feature-classifier:

  No matches found

Debug info has been saved to /var/folders/gt/2v1g_2ld469c8hxyxctlt4xc0000gn/T/qiime2-q2cli-err-vpauoyiz.log

Note that the “correct_no_spaces_only_bifido_and_gardnerella_cpn_rep_seqs_upper_no_replicates.qza” is the fasta file of my reference database and the primer sequences are the primers I use to amplify the groEL gene not the indices. So I am unsure of what the error is. Do you know what the issue could be? Thank you!

[thermokarst]

Can you please rerun with the --verbose flag and paste the complete error message here? Thanks!

[Stephanieorch (Stephanie Orchanian)]

[1]+ Stopped qiime feature-classifier extract-reads --i-sequences correct_no_spaces_only_bifido_and_gardnerella_cpn_rep_seqs_upper_no_replicates.qza --p-f-primer TCGTCGGCAGCGTCAGATGTGTATAAGAGACAGTCCGATTACGAYCGYGAGAAGCT --p-r-primer GTCTCGTGGGCTCGGAGATGTGTATAAGAGACAGCSGCYTCGGTSGTCAGGAACAG --p-trunc-len 490 --o-reads ref-seqs-cpn.qza
(qiime2-2019.1) Stephanies-MacBook-Pro:Downloads stephanieorch$ qiime feature-classifier extract-reads --verbose --i-sequences correct_no_spaces_only_bifido_and_gardnerella_cpn_rep_seqs_upper_no_replicates.qza --p-f-primer TCGTCGGCAGCGTCAGATGTGTATAAGAGACAGTCCGATTACGAYCGYGAGAAGCT --p-r-primer GTCTCGTGGGCTCGGAGATGTGTATAAGAGACAGCSGCYTCGGTSGTCAGGAACAG --p-trunc-len 490 --o-reads ref-seqs-cpn.qza
Traceback (most recent call last):
File “/Users/stephanieorch/miniconda3/envs/qiime2-2019.1/lib/python3.6/site-packages/q2cli/commands.py”, line 274, in call
results = action(**arguments)
File “</Users/stephanieorch/miniconda3/envs/qiime2-2019.1/lib/python3.6/site-packages/decorator.py:decorator-gen-342>”, line 2, in extract_reads
File “/Users/stephanieorch/miniconda3/envs/qiime2-2019.1/lib/python3.6/site-packages/qiime2/sdk/action.py”, line 231, in bound_callable
output_types, provenance)
File “/Users/stephanieorch/miniconda3/envs/qiime2-2019.1/lib/python3.6/site-packages/qiime2/sdk/action.py”, line 365, in callable_executor
output_views = self._callable(**view_args)
File “/Users/stephanieorch/miniconda3/envs/qiime2-2019.1/lib/python3.6/site-packages/q2_feature_classifier/_cutter.py”, line 168, in extract_reads
raise RuntimeError(‘No matches found’) from None
RuntimeError: No matches found

[Nicholas_Bokulich]

Hi @Stephanieorch,
Those primers look very long, and I am guessing you are including adapter sequences in there. See here for more details:

Plugin error: feature-classifier

Hi @Cheng50373640, It looks like the primers you are inputting are the full tailed PCR primers provided in, e.g., Nextera kits. These primer sequences contain Illumina adapter sequences on the 5’ end that do not anneal to target DNA, and do not match any known sequences in nature (they are designed this way to avoid amplification bias and other issues). Hence, the 5’ portion of these sequences will not match any sequences in Greengenes, resulting in a lack of hits. What you need to provide to e…

Let us know if that makes sense or if your primers are really that long!

[Stephanieorch (Stephanie Orchanian)]

The adapter sequences are not included in there, I’m just using long primers.

[Nicholas_Bokulich]

@Stephanieorch

The adapter sequences are not included in there, I’m just using long primers.

Then there must simply be no matches for that primer in the reference database that you are using.

You can attempt to spot check a few sequences to confirm, but for what it's worth an NCBI BLAST search using the forward primer does not retrieve any good hits in the nt database.