[mira_talk] Re: Quality values for Illumina

  • From: Bastien Chevreux <bach@xxxxxxxxxxxx>
  • To: mira_talk@xxxxxxxxxxxxx
  • Date: Thu, 14 Jul 2011 00:18:50 +0200

On Jul 13, 2011, at 23:33 , Peter Cock wrote:
> On Wednesday, July 13, 2011, Arun Rawat <rawat_arun@xxxxxxxxx> wrote:
>> The new Illumina software outputs the quality value in offset of 33 ASCII as
>> against 84 ASCII offset. Do you think it will affect my cleaning quality if I
>> give following CL:qc=yes:qcmq=5:qcwl=5 and what might be the solution for 
>> this?
> For the record, you have a typo - the old offset was 64.
> Note for Illumina 1.8 the new offset is 33 as per the original Sanger
> FASTQ standard.
> I think it will be fine as MIRA tries to handle Sanfer/Solexa/Illumina
> FASTQ automatically, but there is a switch to set this explicitly if I
> recall correctly.

I'm getting since quite some time data from my provider with 33 as offset, as 
well as the SRA has them in that format. So MIRA now has that also as default, 
but it should also handle offset 64 data fine by default.

What will not work is FASTQ offset 64 and old scoring scheme with "negative" 
quality scores ... I never got that kind of data and I think it was ephemeral 
enough to simply ask people to do the conversion with other tools.

The switch Peter refers to is a very old thing which works only when the data 
is present in FASTA / FASTA-Quality files. For Illumina, these formats too have 
fortunately gone the way of the Dodo and the only reason I keep the switch is 
that I still have a couple of old reference data sets I need to work on fro 
time to time.

B.


--
You have received this mail because you are subscribed to the mira_talk mailing 
list. For information on how to subscribe or unsubscribe, please visit 
http://www.chevreux.org/mira_mailinglists.html

Other related posts: