missing index.fastq in MiSeq output

[alexis rapin]

Hello,

I have been sequencing a 16S library on a MiSeq but forgot to add "<add key="CreateFastqForIndexReads" value="1" />" in the MiSeq Reporter.exe.config file. Thus I only have R1.fastq and R2.fastq as output (both are not demultiplexed and do not contain any index).
Does anybody know how to generate the index.fastq post-run ?

I know there is the bcl2fastq software available but it is supposed to convert and demultiplex in the same time.
I just want to create index.fastq and have identifiers linking sequences from R1.fastq, R2.fatsq and index.fatsq so that I can demultiplex using split_libraries_fastq.

Thank you,

Alexis

[Tony Walters]

Alexis, were your primer constructs designed in such a way that you need to have three reads to get the barcode/index read (like figure 1 of http://www.ncbi.nlm.nih.gov/pubmed/20534432)?

If the barcodes are at the beginning of the the read 1 (and/or 2), there is a way to create a separate barcodes file with the extract_barcodes.py script (http://qiime.org/scripts/extract_barcodes.html). The resulting barcodes and reads generated with extract_barcodes.py can then be used with split_libraries_fastq.py as normal.

--

---
You received this message because you are subscribed to the Google Groups "Qiime Forum" group.
To unsubscribe from this group and stop receiving emails from it, send an email to qiime-forum...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[alexis rapin]

Thank you very much for helping.

Yes, the construct is similar to Fig1 and requires 3 reads. I only got R1.fastq and R2.fastq and I need Index.fastq.
The sequencer did 3 reads (with R1 sequencing primer, index sequencing primer and R2 sequencing primer). However, it did not give any automatic FASTQ output for the Index read (forgot to modify the config file for this).
I guess it is possible to build Index.fastq from some BCL files just like R1.fastq and R2.fastq were generated but I am a bit lost and don't want to restart a new 1500$ run just to get the index.fastq generated automatically...

[Tony Walters]

Hello Alexis,

I unfortunately do not have direct experience with doing such conversions. It does look like there is software from Illumina that converts bcl to fastq (bcl2fastq): http://supportres.illumina.com/documents/myillumina/ceb38810-038f-410a-b0d6-db0c1f774b28/bcl2fastq_userguide_15038058a.pdf

You would have to install the software, if it wasn't there already, and use the option that does not do demultiplexing.

--

[H W]

It sounds like the easiest way is to perform the demultiplexing manually with Bcl2Fastq which is contained in Illumina's CASAVA pipeline (http://support.illumina.com/sequencing/sequencing_software/casava/downloads.ilmn) Not that the sample sheet for demultiplexing is not the same sample sheet that is used when starting a MiSeq run.
(I suppose the index sequences are not in the fastq-file headers?)

[alexis rapin]

Worked well with Bcl2Fastq!

In the run output directory I used Bcl2Fastq as following and it nicely generated FASTQ for the Index read without demultiplexing:
/usr/local/bin/configureBclToFastq.pl --input-dir Data/Intensities/BaseCalls --output-dir bcl2fastq_output --sample-sheet sample_sheet.csv --no-eamss --use-bases-mask Y251,Y12,Y251 --force
(Used a sample sheet as described in the user's manual with only one sample and a fake 12b barcode)

Thanks for helping!