[Bioperl-l] SeqIO::abi use

[Phillip San Miguel]

Hi Tom and Roy,
    Not sure what the intention is here, but there are caveats you might 
want to consider:
    The problem with looking at the raw trace data is that it has not 
yet been base called, no peaks have been found, nor have the dye-shifts 
been compensated for, nor has any background been subtracted. So finding 
the peaks in the raw data that correspond to the mixed base peak of 
interest is not trivial. Then once you have, you would need to subtract 
background (which can be different for each of the data channels.) Then 
integrate the peaks.
    Then what do you have? The % of the base at that position is A,C,G 
or T?
    I would not be so sure. The dyes may give different quantum yields, 
the CCD may be more sensitive to some wavelengths than others, the 
sequencing polymerase may prefer to add a certain base in your given 
sequence context more than another. Really you would need to do standard 
runs (mixing sequences in each of the base channels to be analyzed of 
each base) in known ratios and create a standard curve to be sure.

Phillip
    
Roy Chaudhuri wrote:
> Hi Tom,
>
> It looks from the docs/source like you can, with something like this:
>
> use Bio::SeqIO;
> my $seq=Bio::SeqIO->new(-file=>$ARGV[0], -format=>'abi', 
> -get_trace_data=>1)->next_seq;
> my @atrace=$seq->get_trace_graph(-trace=>'a'); # or c,g,t
>
> Sorry, but I don't have Staden installed so I can't test if this works.
> You might also look at the non-BioPerl ABI module on CPAN - again, I 
> haven't tested this, but it looks like it could be useful for your 
> purposes:
> http://search.cpan.org/~MALAY/ABI/ABI.pm
>
> Roy.
>
>
> On 08/04/2010 23:30, Tom Keller wrote:
>> Greetings, Can the Bio::SeqIO::abi module be used to access the raw
>> fluorescence data? Are there any examples available? I'm wondering
>> how to calculate ratios of mixed bases prior to the smoothing and
>> scaling done by the base caller.
>>
>> thank you Tom kellert at ohsu.edu<mailto:kellert at ohsu.edu> 503-494-2442
>>
>>
>>
>>
>>
>>
>>
>>
>> _______________________________________________ Bioperl-l mailing
>> list Bioperl-l at lists.open-bio.org
>> http://lists.open-bio.org/mailman/listinfo/bioperl-l
>
> _______________________________________________
> Bioperl-l mailing list
> Bioperl-l at lists.open-bio.org
> http://lists.open-bio.org/mailman/listinfo/bioperl-l
>