Relatively normal percentage passed filter for all but one sample

[jalen_bailey]

Hello! I'm running DADA2 on 5 fastq files. 4 of which are sewage waste, and one of tap water. There's an odd situation where a lot of the reads in my tap water sample did not pass the filter. These are paired-end demultiplexed samples sequenced by Illumina. According to other forum posts, a passing passing percentage this low is a cause for concern however I wouldn't even know where to start in terms of diagnostics. The Parameters of DADA2 are as shown below:

qiime dada2 denoise-paired
--i-demultiplexed-seqs demux.qza
--p-trunc-len-f 294
--p-trunc-len-r 223
--o-representative-sequences rep-seq.qza
--o-table table.qza
--o-denoising-stats denoising-stats.qza

Here's the demux summary and the denoising stats too.

demux-paired-summary.qzv (318.8 KB)

statsvisual.qzv (1.2 MB)

Thankyou for any help!

[colinbrislawn (Colin J Brislawn)]

Hello Jalen,

Welcome to the forums!

Thanks for posting your full command and output files.

The first thing that jumped out to me was the small number of reads in your water sample.

It makes sense that sewage has fewer reads than tap water because sewage probably has more biomass than top water. (Hopefully! )

Low-biomass samples can be strange. With low biomass, low 16S copies, and low reads, it can hard to tell if a sample failed to sequence well, or if nothing was in it to start with!

Perhaps if the cohorts were balanced (5 sewage and 5 tap water), this would tell us more.

[jalen_bailey]

Hello Colin,

Thankyou for the input! I was originally worried that it may have been something with my parameters but I'm happy that it wasn't a problem with that. I'll keep a balanced cohort in mind, the next time we take samples