Taxonomic composition gives errors, Archaea rather the bacteria

[Jawairia]

I am using qiime 2 for taxonomic classification and while extracting seqs (relevant to my primers) from reference data base (SILVE132_99_OTUS) and further downstream analysis with classifier, I only get Archeae in all of the samples. I tried with many things but no success. Do you know why it looks like this? I also run samples that I have run with Qiime 1 last year and gave me the good taxonomic classification, but in qiime 2 they also gave Archaea only. its strange. may be I am missing something or doing something wrong.

urgent help required.

1. qiime feature-classifier extract-reads --i-sequences silva132_99_otus.qza --o-reads ref-seqs.qza --p-f-primer. (my primers) --p-r-primer (my primers)
2. qiime feature-classifier fit-classifier-naive-bayes --i-reference-reads ref-seqs.qza --i-reference-taxonomy ref-taxonomy.qza --o-classifier classifier.qza --verbose
3. qiime feature-classifier classify-sklearn --i-classifier classifier.qza --i-reads rep_seqs_97_nonchimeras.qza --o-classification taxonomy.qza --p-n-jobs 16
4. qiime metadata tabulate --m-input-file taxonomy.qza --o-visualization taxonomy.qzv

regards,

Jawairia

[Nicholas_Bokulich]

Hi @Jawairia,
Please see this post. It sounds like you have the same issue as the others in that thread.

As a workaround, I recommend A) removing that sequence, B) increase the --p-identity parameter (e.g., to 0.9) for extract-reads, or C) do not do the read extraction (this boosts accuracy by a bit compared to full-length, but it is not a game-changer).

I hope that helps!

[Jawairia]

Thanks, I will do the B do not extract reads. and will see how does it works. and also read the posts.

I will get back once out will be fixed or give new issues.

Best,

Jawairia