Demultiplexing slow

[Sivakumar Gowrisankar]

Hey Everyone

I am experiencing terribly slow demultiplexing operation when using CASAVA 1.7 demultiplexer (24 hrs). It runs fairly faster on an another cluster that I don't have many reserved nodes (1-2 hrs). Has anyone encountered a similar problem and if so do is there any software issues that could be causing this ?

The cluster that is causing the slow demultiplexing has better configuration (8core, 16G) compared to the one that has faster demultiplexing (1core, 4G).

Please let me know if you have any inputs.

Siva
PCPGM

[DavidJ]

I'm using "index_decoder" from http://sequenceread.sourceforge.net/
It's resonably quick (e.g 5 1/2 minutes for 1 lane of 15,581,707
clusters tagging on 4 different 8 base tag sequences - suspect NFS io
was limiting factor). Works on qseqs and fills in the index column/
field. Don't think it can separate them to different files.

[Sivakumar Gowrisankar]

Thanks for sharing this David. After a lot of communication with Illumina, it has come up with a patch that effectively fixes this io issue. Now it takes about 10-15 minutes 25 million reads.

Siva

--
You received this message because you are subscribed to the Google Groups "solexa" group.
To post to this group, send email to sol...@googlegroups.com.
To unsubscribe from this group, send email to solexa+un...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/solexa?hl=en.

[Kevin M. Carr]

index_decoder is just the first step in the process if you want
separate files. The sequenceread package provides two additional
utilities (granted for doing more than just demultiplexing). After
index_decoder you would run illumina2srf using the indexed qseq files
as input to create a single SRF file. Then you would run
split_srf_by_tag to create individual SRF files for each tag.

From the individual SRF files you output separate FASTQ files using
srf2fastq (part of the io_lib package, a dependency for the
sequenceread package).

Kevin M. Carr

**************************
Bioinformatics Specialist
Research Technology
Support Facility
S-20 A Plant Biology Lab
Michigan State University
East Lansing, MI 48824

Ph: (517) 355-6759 x102
Fax:(517) 355-6758
**************************