format of AB SOLiD 4 System sequencing output

[christear]

To validate one of my hypothesis, I've downloaded some public data from EMBL-EBI ENA (European Nucleotide Archive) (http://www.ebi.ac.uk/ena/).
The data is from a paper published in Nature structural & molecular biology in 2011. It was generated by AB SOLID 4 System.
As described in the ENA for this data set, the Fastq files are available both via ftp or galaxy.
The problem is that , I found that the fastq file that I downloaded is so wired
and I have never faced this before. Details are showed as following.

###Eg. 09_public_data$ less ERR042386.fastq

@ERR042386.1 solid0032_385_1_4_20100830_FRAG
T32120132000132211310023202201202002303130332322311
+
!@%62B8?=A690@>><->8=51%:==5521=582<@9>9><,6785.>4&

Generally, in a classic fastq format file, first line is begin with "@", 2nd line is the sequence of reads, 3rd line is a "+" and 4th line is the quality.
However in these fastq files, the sequence of reads are some numbers ("0,1,2,3"). I really have no idea what does it means ...

Is that ("0,1,2,3") represent ("A,G,C,T") respectively ?
or is it a unique format for ABI solid sequence output format ?

Does someone have experience to deal with this kind of data ?
All suggestions are appreciated ...

[mastal]

format of AB SOLiD 4 System sequencing output

It's a unique format for SOLiD, what you're seeing is the sequence in colorspace.

SOLiD uses a dibase encoding system, where each color represents a sequence of two bases.

Have a look at some of the manuals on the Life Technologies website,

http://www.appliedbiosystems.com/absite/us/en/home/applications-technologies/solid-next-generation-sequencing/next-generation-systems/solid-sequencing-chemistry.printable.html

[christear]

Thanks a lot ...
I know SOLID using dinucleotides enconding the sequence.
However, what i downloaded is already fastq format file, at least it should be converted to AGCT...
I have analyzed solid data before whereas it's the normal sequence in Fastq format file...
Anyway, thanks a lot ... Do you know some tools to do the convention?

[JPC]

With Solid data you need to do the mapping in base-space, not convert the fastq and map that. If you're not familiar with it I would recommend tracking down an expert.

I don't know if it is still available but I think the Life Tech software was called BioScope, they now have 'LifeScope' but I don't know if that is good for v4 machines.