[Biojava-l] About ABITrace class
Andy Yates
ayates at ebi.ac.uk
Mon Jan 7 09:43:24 UTC 2008
Yup Russ is on the ball with that one 100% :)
A chromatogram is really a graph of 4 tracks where each scan index
refers to an intensity of the 4 dyes used in Sanger Sequencing. There
is no direct relation to the size of a chromatogram and the number of
base pairs it will represent. If you require any changes to the
rendering the options are controlled by ChromatogramGraphic.Option
instances in conjunction with the setOption() method in
ChromatogramGraphic.
If you are rendering a processed ABI file and just want to see the
base called sequences then the best option is to use the
Chromatogram's getBaseCalls() method & look up the start & finish of
the base calls. Use this to set where you want rendering to start &
stop (again another ChromatogramGraphic.Option).
Hope this helps,
Andy
On 30 Dec 2007, at 17:04, Russ Kepler wrote:
> On Friday 28 December 2007 01:54:30 David Bourgais wrote:
>> Hello BioJava users
>>
>> I am writing a little program using BioJava 1.5 and JFreeChart 1.0.8.
>> My aim is to display with JFreeChart a chromatogram by reading an ab1
>> file with the BioJava help.
>> Okay, I can display my chromatogram. But, my chromatogram size is
>> 13301
>> (using getTrace(AtomicSymbol base)) when my sequence size is 1110 bp
>> (according to the the BufferedImage produced with BioJava).
>> Why this difference ? How can I correct my program in order to see
>> correctly my chromatogram ?
>
> If I understand you correctly you're looking at the trace data
> (multiple
> points per peak) and comparing to the base calls (one call per
> peak). In my
> experience folks interested in the chromatogram would prefer to see
> an image
> with the chromatogram spaced properly, those primarily interested in
> the base
> calls prefer to have the chromatogram scaled around a fixed width
> for the
> basecall. There's code to do either in BioJava, but the former
> might not be
> setup to generate a BufferedImage (if you can't find it I might be
> able to
> help some).
> _______________________________________________
> Biojava-l mailing list - Biojava-l at lists.open-bio.org
> http://lists.open-bio.org/mailman/listinfo/biojava-l
More information about the Biojava-l
mailing list