Qiime2 Demux Error

Question

[zach]

Hello there,

I hope there is someone experienced in Qiime2 who could help me please. I am demultiplexing sequence data with the command below:

qiime demux emp-paired   –m-barcodes-file 54386_mapping_file.txt –m-barcodes-column BarcodeSequence   –p-rev-comp-mapping-barcodes –i-seqs emp-paired-end-sequences.qza    –o-per-sample-sequences demux.qza  –o-error-correction-details demux-details.qza

But I received the error: Plugin error from demux: Barcode header lines do not contain description fields but sequence header lines do.

Although the barcodes.fast.gz is in proper 4 line format, I noticed the 4th line is '@@@@' for the 1st few lines which could have been the potential issue:

@DGZN8DQ1:549:H7C23BCXX:2:1101:1087:1870
CGTCGTATGAAT
+
@@@@@@@@@@@@

I then used 'seqtk' to change the 4th line to '####' to produce these 1st few lines:

@DGZN8DQ1:549:H7C23BCXX:2:1101:1087:1870
CGTCGTATGAAT
+
############

However, I again received the same error about description fields in the sequence but not barcode header lines. How do I identify this problem and rectify it to demux successfully? Thank you in advance for any help!

[GenoMax]

Although the barcodes.fast.gz is in proper 4 line format, I noticed the 4th line is '@@@@' for the 1st few lines which could have been the potential issue:

@ is a valid quality score (Q31) for sanger formatted fastq files.

[zach]

I also noticed another potential and more likely problem. The sequence header is this:

@DGZN8DQ1:549:H7C23BCXX:2:1101:1087:1870 1:N:0:CGTCGTATGAAT

Whereas the barcode header is this:

@DGZN8DQ1:549:H7C23BCXX:2:1101:1087:1870

I think the error is probably because the barcode header lines do not contain the 'description field' of 1:N:0:CGTCGTATGAAT

If this is the problem, is there a way I could add the corresponding description fields to all my barcode headers? Thanks and apologies for the long question.

[GenoMax]

That is not description field. It is the sequence of the index which is needed to demultiplex the data.

So you have a separate file for the index sequences. You can use this solution to demultiplex the data: A: Demultiplexing Illumina data

[zach]

Thanks for all the answers, genomax. From the link you provided, I am trying to demux with deML as it seems like a promising solution. If I successfully get the output in fastq format, is there a way to convert the demux-ed document to .qza format? I need it in .qza for Qiime2 donwstream analyses.

I have another question slightly related. If demux-ing with Qiime2, would it work if I simply add the corresponding sequence of the index to every barcode header? If yes, how could I do this with the header of a separate sequence file? I guess it's a similar thing you did with your reply in this forum discussion - http://seqanswers.com/forums/showthread.php?t=74570

eg. Add '1:N:0:CGTCGTATGAAT' to barcode header and the same for other unique samples.

I hope you understand my questions. Thank you again!

[GenoMax]

You need to actually run Qiime2 analysis to get the .qza files.

As for your second question it may. Qiime2 evolved over years to use different inputs. Take a look at the input section in the tutorial for more.

[zach]

Regarding my 2nd question, is there a way I could add the sequence of the index (from forward sequence file) to the barcode file headers for each sample?

I am quite new to the Linux system. I am wondering how I could possibly utilise the 'paste' command (?) to solve this problem? I would be grateful if you could help me with this!