Analyze Sanger Sequences for 23S

Hi!!
I made a clone library to amplified the Domain V of the 23S rRNA with the primers p23SrV_f1 and p23SrV_r1. After Sanger sequencing I got 420 sequences (14 samples, 30 reads for each sample), and I want to analyze them to create SV’s instead of OTU’s.
First, I created a classfier for the Silva LSU 132 database following the steps that is included in the .txt file for the Silva SSU 132
Then, I started to process the raw data as following:

  • converting the ab1 files into fastq files with SeqIO
  • find which sequences are in forward and reverse direction with usearch -search-oligodb
  • trim the vector and primers by length with usearch -fastx_truncate getting sequences of 363 n
  • reverse complemented the reverse sequences with useach -fastx_revcomp
  • change the quality max score from 62 to 41 with vsearch --fastq_convert
    at this point I have 14 fastq files, one for each sample

In qiime2:

  • I imported the sequences with the fastq manifest format SingleEndFastqManifestPhred33
  • tried to denoise with DADA2

qiime dada2 denoise-single
–i-demultiplexed-seqs single-end-demux.qza
–p-trim-left 0
–p-trunc-len 0
–p-chimera-method none
–o-representative-sequences rep-seqs-dada2.qza
–o-table table-dada2.qza
–o-denoising-stats stats-dada2.qza

  • I think DADA2 find the singletons as noise and discarded them

the stats are: https://view.qiime2.org/visualization/?type=html&src=https%3A%2F%2Fdl.dropbox.com%2Fs%2Fvd2ihc7gnpyhh63%2Fstats-dada2.qzv%3Fdl%3D1

  • the represent sequences resulting only in 23

https://view.qiime2.org/visualization/?type=html&src=https%3A%2F%2Fdl.dropbox.com%2Fs%2Fsxp63va4cv2da7g%2Frep-seqs-dada2.qzv%3Fdl%3D1

Finally I ran the taxonomic classification and I got:

https://view.qiime2.org/visualization/?type=html&src=https%3A%2F%2Fdl.dropbox.com%2Fs%2Fslpiq8pq3xnhrx6%2Ftaxa-bar-plots.qzv%3Fdl%3D1

I wanted to denoise with deblur but I dont know where or how to get the reference database for the LSU

So, I want to know if you have any suggestion
Should I analyze my sequences with the OTU approach?

deblur and dada2 are both sequencing platform-specific so (as far as I know) would not be appropriate for Sanger sequence data.

Instead you can just dereplicate these sequences (essentially creating 100% OTUs).

that is correct

this would not be an issue with dereplicating with q2-vsearch

I hope that helps!

Thanks for your reply!!

Another question, the classifier is platform-specific? Or could be useful?

Good question. No, the classifier is not platform specific so any of the classification methods it should be useful for Sanger sequences.

This topic was automatically closed 31 days after the last reply. New replies are no longer allowed.