Demultiplexing nested libraries

Hi all,

I am struggling with something that I ought not to be struggling with. I've had a good look around the forums and I haven't seen an identical issue so here goes.

I've been given a paired end 16S EMP MiSeq run. The researcher took a nested barcode approach. 10 libraries were prepared, where each individual library has 25 samples in. Each of these 25 samples has been tagged with forward and reverse barcodes, and the 10 libraries were then tagged again.

I've acquired the 10 PE FASTQ files, demultiplexed from BaseSpace. My aim is now to treat these 10 libraries as single multiplexed libraries, and demultiplex them all to get my 250 individual samples.

To do this, I have been using QIIME2 and following the EMPPairedEndSequences import steps. I can get a multiplexed QZA file for a library, comprised of forward.fastq, reverse.fastq, and barcodes.fastq.

Next, I wish to use qiime demux emp-paired to demultiplex my QZA. However, the researcher has reused forward barcodes, mixed with different reverse barcodes. Therefore, I have duplicate forward barcodes, but these have different reverse barcodes. I cannot input multiple columns into --m-barcodes-column.

I feel like the solution to this is really obvious, but it's got me totally stumped! Any help or advice will be gratefully received!

Thanks,
Peter

Hi @peterleary,
So if I understand this correctly, at the current set up you have 10 separate multiplexed .qza files that you are trying to demultiplex. These libraries are dual-indexed with a forward and reverse barcodes. The issue then comes down to the fact that you are trying to demultiplexing using the demux emp-paired plugin which is specific to the EMP protocol which assumes there are only barcodes on the forward reads. Instead, try using cutadapt demux-paired which allows demultiplexing of dual barcodes. Note that you'll have to make an additional barcode column in your metadata file.
Hope I understood this right and this helps.

2 Likes

Hi Mehrbod,

Thank you so much, that's worked perfectly. I knew the answer was really obvious!

For future reference, here's what I did:

  • Put the forward and reverse FASTQ files for library 1 into a directory, and renamed the files forward.fastq.gz and reverse.fastq.gz

  • Ran the following command:

qiime tools import \
--type 'MultiplexedPairedEndBarcodeInSequence' \
--input-path fastq_lib1 --output-path demux/L1.qza

  • Then the following command:

qiime cutadapt demux-paired \
--i-seqs demux/L1.qza \
--m-forward-barcodes-file L1.metadata.txt \
--m-forward-barcodes-column BarcodeSequence \
--m-reverse-barcodes-file L1.metadata.txt \
--m-reverse-barcodes-column ReversePrimer \
--o-per-sample-sequences demux/L1.demux.qza \
--o-untrimmed-sequences demux/l1.untrimmed.qza

  • Then I used qiime demux summarise and qiime tools export commands.

Thanks again!
Peter

1 Like

HI @peterleary,
Glad it worked out and thanks for following up with your solution. Really appreciate it!

1 Like

This topic was automatically closed 31 days after the last reply. New replies are no longer allowed.