Dada2 in Qiime2: losing reads during merging

Question

[kamanovae]

Hi!

I am trying to use dada2 on my 16S paired-end sequences, from a 2 x 150 bp illumina run. But I am losing almost all my reads at the merging step. The FASTQC shows excellent data quality and no adapters in all my samples.

My pipeline looks like this:

1. qiime tools import --type 'SampleData[PairedEndSequencesWithQuality]' --input-path manifest.tsv --output-path pe-demux.qza --input-format PairedEndFastqManifestPhred33V2
2. qiime dada2 denoise-paired --i-demultiplexed-seqs pe-demux.qza --p-trim-left-f 15 --p-trunc-len-f 130 --p-trunc-len-r 120 --p-n-threads 20 --o-table table.qza --o-representative-sequences rep-seqs.qza --o-denoising-stats denoising-stats.qza &

Fastqc don't find primers, but I tried p-trim-left parameter in dada2. But still see the same problem of loss of reads during merging. Finally, all my attempts to run the program give the same result. It is shown in the screenshot. I have only 0-100 merged reads, and only <0.1 of them is non-chimeric. But % of input passed filtered is >95%

I received fast sequencing data and don't know what adapters were used to prepare the Illumina library.

After unsuccessful attempts with the dada2 program, I tried to use deblur.

1. qiime tools import --type 'SampleData[PairedEndSequencesWithQuality]' --input-path manifest.tsv --output-path pe-demux.qza --input-format PairedEndFastqManifestPhred33V2
2. qiime vsearch join-pairs --i-demultiplexed-seqs ../pe-demux.qza --o-joined-sequences demux-joined.qza
3. qiime quality-filter q-score --i-demux demux-joined.qza --o-filtered-sequences demux-joined-filtered.qza --o-filter-stats demux-joined-filter-stats.qza
4. qiime deblur denoise-16S --i-demultiplexed-seqs demux-joined-filtered.qza --p-trim-length 140 --p-sample-stats --p-jobs-to-start 4 --o-representative-sequences rep-seqs.qza --o-table table.qza --o-stats deblur-stats.qza

As a result, I got even worse results than dada2

I would be glad for any advice on merging my readings

[andres.firrincieli]

How long was the amplicon?

If the amplicon is longer than the sum of the two read lengths, the R1 and R2 reads will not merge. By default, dada2 need a 10 bp overlap for merging

[kamanovae]

Thanks for the answer!

I requested information from the lab. Indeed, a region was sequenced that does not overlap with such short reads of 150 bp. Can I parse forward and backward reads as SingleEndFastqManifestPhred33V2. But how do I then combine the results? I want to get a general taxonomic table and graphs for alpha diversity.

[andres.firrincieli]

Honestly, I would proceed only with the R1 reads however, you should probably post this question in the qiime2 forum. You can probably find more experts there