Moving pictures tutorial: Demultiplexing

Hi everybody,
I’m new to QIIME2, and I have a question regarding demultiplexing in moving pictures tutorial. I had a look at barcodes.fastq.gz which contains the barcode read associated with each sequence in sequences.fastq.gz, and I tried to allocate the barcodes inside the sequences, but I couldn’t. Any idea about where exactly are the barcodes within the sequences? Thanks.

Hi, @Rania,

Welcome to the :qiime2: forum!

Can you elaborate on what you mean by trying to allocate the barcodes inside the sequences? What exactly did you try? What were the expected results vs actual results?

I think the observations noted below will help clarify some things:

If you compare the sequence identifiers in both sequences.fastq and barcodes.fastq emp-single-end-sequences/sequences.fastq.gz and emp-single-end-sequences/barcodes.fastq.gz from the moving pictures tutorial), you will notice that the identifiers match up. I'll demonstrate this with the command below which simply displays the first entry in both sequences.fastq and barcodes.fastq.

➜ head -n4 sequences.fastq barcodes.fastq
==> sequences.fastq <==
@HWI-EAS440_0386:1:23:17547:1423#0/1
TACGNAGGATCCGAGCGTTATCCGGATTTATTGGGTTTAAAGGGAGCGTAGATGGATGTTTAAGTCAGTTGTGAAAGTTTGCGGCTCAACCGTAAAATTGCAGTTGATACTGGATATCTTGAGTGCAGTTGAGGCAGGGGGGGATTGGTGTG
+
IIIE)EEEEEEEEGFIIGIIIHIHHGIIIGIIHHHGIIHGHEGDGIFIGEHGIHHGHHGHHGGHEEGHEGGEHEBBHBBEEDCEDDD>[email protected]@B>@@@@@[email protected]@@[email protected]@=>?08;3=;==8:5;@6?######################

==> barcodes.fastq <==
@HWI-EAS440_0386:1:23:17547:1423#0/1
ATGCAGCTCAGT
+
IIIIIIIIIIIH

In this case, the identifier @HWI-EAS440_0386:1:23:17547:1423#0/1 appears in both sequences.fastq and barcodes.fastq. Note that you need to unzip the fastq files before you can look at them.

You will also notice a 1-1 correspondence between the sequences in barcodes.fastq and values in the barcode-sequence column of sample-metadata.tsv.

➜ grep ATGCAGCTCAGT qiime2-moving-pictures-tutorial/sample-metadata.tsv
L2S204	ATGCAGCTCAGT	left palm	2009	3	17	subject-1	No	140

I hope this helps! :nerd_face:

Hi @andrewsanchez,

Thank you so much for your reply. That was really useful. My problem was that I thought that the barcodes were still in the sequences, so I tried to find the barcodes within the reads. But it turned out that they have already been removed. That’s why q2-demux is used in this tutorial, otherwise cutadapt would be used instead.

Regards,
Rania

2 Likes

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