DADA2 Write output Error

Technical Support
1

Dear all,

I have sequenced samples I need to run with QIIME2. I use 2019.4 but this error always showing (attached). This is the command line used

nohup qiime dada2 denoise-single \
--i-demultiplexed-seqs single-end-demux.qza \
--o-table feature_table.qza \
--o-representative-sequences feature_seqs.qza \
--p-trunc-len 0 \
--p-n-threads 0 \
--verbose \
--o-denoising-stats dada2_stats.qza > dada2.log &

I could not understand where is the issue.
Appreciate your help

dada2.txt (3.3 KB)

2

Hello Faisal,

Thanks for posting this! This is a very interesting error. I’m also stumped by this, but let’s dive in.

Here are my first two thoughts:
Rerun this command without nohup and > dada2.log &
Rerun this command using the data in the tutorial.

I’m suggesting this because I’m wondering if…
… The extra stuff like nohup and > dada2.log & are confusing qiime
… Something went a little bit wrong curing the conda install, and this is causing an issue.

Let me know what you have found!
Colin

1 Like
3

Another potential option:

It looks like the pandas CSV parser has failed from inconsistent column sizes. Some areas to explore there:

  1. R skipped some data, which implies the denoising stats were “null”. I don’t think I’ve seen this, so it seems unlikely.
  2. R wrote a \ which changes the parsing semantics, this might be a result of special characters or just a literal \.

And of course the install error potential like @colinbrislawn mentioned.

What do your sample IDs look like, and how did you install QIIME 2 (and what version)?

4

Thank you @colinbrislawn and @ebolyen for your reply.

I found the issue with the CSV file that import all sequences as manifest. The last sample had extra tab to its name which for some reason made this issue. I just removed it and everything went fine!

Small mistake can make major issue.

Thanks again!
Faisal

4 Likes