Demux Output 0?

[2Black_Cats]

Hello!

I sent my DNA samples off to a sequencing center for sequencing and had some basic analyses performed by one of their bioinformaticians. I'm trying to walk through what he did so I can understand the code myself, but have run into some issues.

Qiime version: qiime2-2023.5 (installed in a conda environment, and I'm working in Terminal)

The code I've run so far is as follows:

qiime tools import \
  --type "SampleData[SequencesWithQuality]" \
  --input-format SingleEndFastqManifestPhred33V2 \
  --input-path ./manifest.tsv \
  --output-path ./demux_seqs_original.qza

qiime cutadapt trim-single --i-demultiplexed-sequences demux_seqs_original.qza --p-cores 8 --p-minimum-length 240 --o-trimmed-sequences demux_seqs.qza --p-adapter GTGYCAGCMGCCGCGGTAA...ATTAGAWACCCBNGTAGTCC --p-discard-untrimmed

qiime demux summarize --i-data ./demux_seqs.qza --o-visualization ./demux_seqs.qzv

My manifest.tsv file looks like, but for brevity, I'm just listing the first 2:

|sample-id|absolute-filepath|
|M01|/home/<username>/scratch/rawreads_16S/samples/1691-01-M1_S291_L001_R1_001.fastq.gz|
|M02|/home/<username>/scratch/rawreads_16S/samples/1691-02-M2_S292_L001_R1_001.fastq.gz|

demux_seqs.qzv

What else should I try?

[timanix (Timur Yergaliyev)]

Hello!

I would reorder and separate commands:

First, import files:

qiime tools import \
  --type "SampleData[SequencesWithQuality]" \
  --input-format SingleEndFastqManifestPhred33V2 \
  --input-path ./manifest.tsv \
  --output-path ./demux_seqs_original.qza

Then create a visualization to check the data:

qiime demux summarize \
  --i-data ./demux_seqs_original.qza \
  --o-visualization ./demux_seqs_original.qzv

And only after checking sequences info run cutadapt:

qiime cutadapt trim-single \
  --i-demultiplexed-sequences demux_seqs_original.qza \
  --p-cores 8 \
  --p-minimum-length 240 \
  --o-trimmed-sequences demux_seqs.qza \
  --p-adapter GTGYCAGCMGCCGCGGTAA...ATTAGAWACCCBNGTAGTCC \
  --p-discard-untrimmed

qiime demux summarize \
  --i-data ./demux_seqs.qza \
  --o-visualization ./demux_seqs.qzv

If sequences are Ok before cutadapt, I would play with --p-minimum-length and double check --p-adapter since --p-discard-untrimmed option is trashing any sequence with no adapter found in it.

[2Black_Cats]

Thank you so much!
Here is a visualization of the demux_seqs_original.qzv

The minimum scores are mainly influenced by my NTCs.

And here are screenshots of the interactive quality plot:

Based on this, would you recommend making --p-minimum-length 154 in cutadapt instead?

[timanix (Timur Yergaliyev)]

I would disable it at all or set to lower value (150) to get rid of sequences shorter than threshold.