Bcl2fastq avoid addition of barcode to header

Question

[GSAENZDEPIP]

Hello, I am demultiplexing some samples with bcl2fastq and I wonder if it is possible to avoid the addition of the barcode sequence to the header of the reads.

Many thanks, Goren

P.D: In the next steps I will use UMI-tools for adding the UMI sequence to the header, so I need to get rid of the barcode first.

[GenoMax]

I don't think you can (if you are truly demultiplexing the data using bcl2fastq). You will need to do it after the fact.

Answer 1

[Devon Ryan]

UMI-tools is perfectly happy with having the Illumina barcode in the normal spot, it adds UMIs to the read names.

[GSAENZDEPIP]

Okey, I will try UMI-tools without removing the barcode and see if it works. Thanks!

[GSAENZDEPIP]

It worked!! Thank you very much

Answer 2

[swbarnes2]

My fastqs have no barcodes in their header. (bcl2fastq v2.19.403) Only undetermined samples have that.

[GenoMax]

How do you know which index the sample has then?

[swbarnes2]

I trust the sample sheet to have that information correct.

[GenoMax]

I have never seen a bcl2fastq demultiplexed file without the index sequence transferred to the fastq header. Perhaps you are using a bcl2fastq option I am unaware of or are post processing files to remove the index sequence from fastq headers.

[swbarnes2]

Nope. No special options, no post processing. Version 2.19 if I remember right. Indices are included in the undetermined fastq file, but not the sample fastqs.

Actually, I see that it depends on the instrument. I run the same version and command line of bcl2fastq on the Miseq, and I see indices in the headers, but my Hiseq1000 fastq files do not have indices in the headers