[Bioperl-l] Re: Questions about Bio::AlignIO::maf

James Thompson tex at biocompute.net
Sat Jan 1 02:32:37 EST 2005


Alison,

You're right on this, I just committed the fix to maf.pm. This also fixed
a range problem in the AlignIO.t test script and the associated humor.maf
test data file, I just committed fixes for those as well.

Thanks for the fix. :)

Cheers,

James Thompson

On Thu, 30 Dec 2004, Lee Ping Alison wrote:

> Hi Mr Thompson,
> 
> Thanks for the reply. I understand the need for the one-based inclusive
> coordinate system now; also partly because the major genome browsers use
> that. However, since you're using inclusive coords, then shouldn't you add 1
> to $start first before calculating $end, since $start is zero-based?
> 
> Alison.
> 
> ----- Original Message -----
> From: "James Thompson" <tex at biocompute.net>
> To: "Lee Ping Alison" <g0404203 at nus.edu.sg>
> Cc: "Allen Day" <allenday at ucla.edu>; "Bioperl" <bioperl-l at bioperl.org>
> Sent: Wednesday, December 29, 2004 3:30 PM
> Subject: Re: [Bioperl-l] Re: Questions about Bio::AlignIO::maf
> 
> 
> > Alison (and Allen),
> >
> > I was the aforementioned bug fixer. :)
> >
> > Sorry if there's any confusion on this, but AFAIK Bioperl uses an
> one-based
> > inclusive coordinate system. While maf may have its own opinions on the
> best
> > way to do coordinates, maf is only one of the formats that are supported
> by
> > Bio::AlignIO.  The consensus in Bioperl appears to be that it makes more
> sense
> > to use one consistent coordinate system within all of the modules rather
> than
> > catering to the opinions and idiosyncrasies of all of the possible file
> > formats. If we did not fix the off-by-one bug in maf.pm, then would be
> > consistency issues with Bio::Align::AlignI objects created from different
> file
> > formats.
> >
> > Here's a link to a message from the mailing list that seems relevant to
> the
> > topic at hand:
> >
> > http://bioperl.org/pipermail/bioperl-l/2002-June/008309.html
> >
> > Cheers,
> >
> > James Thompson
> >
> > On Wed, 29 Dec 2004, Lee Ping Alison wrote:
> >
> > > Hi,
> > >
> > > Mr Day, thanks a lot for helping me with my queries.
> > >
> > > I've just obtained the most recent bioperl-live code via cvs with the
> bug
> > > fixes you've mentioned. I'm wondering why the off-by-one bug fix (end =
> > > start+size-1) was necessary. I'm thinking that "end = start+size" is
> correct.
> > > Because the MAF file format by UCSC states that coordinates are
> half-open,
> > > zero-based. And I have understood it as the coordinates in "maf" module
> > > should be (start, end] (start exclusive, end inclusive). I've also tried
> > > several coordinates that agree with UCSC Genome Browser which uses
> [start,
> > > end]. Hence, in my opinion the bug fix was not necessary.
> > >
> > > Will someone please enlighten me on this?
> > >
> > > Thank you very much!
> > >
> > > Alison.
> > >
> > >   ----- Original Message -----
> > >   From: Allen Day
> > >   To: Lee Ping Alison
> > >   Cc: Bioperl
> > >   Sent: 29 December, 2004 3:34 PM
> > >   Subject: Re: Questions about Bio::AlignIO::maf
> > >
> > >
> > >   Hi Alison,
> > >
> > >   I did not add strand information as I didn't need it at the time of
> > >   writing.  However, I believe this has come up on list recently and
> someone
> > >   has already patched in strand support, as well as an off-by-one bug in
> my
> > >   code.  Can whoever did these patches recently pipe in?  Thanks.
> > >
> > >   Alison, please keep the bioperl list CCed in your reply.
> > >
> > >   -Allen
> > >
> > >   On Wed, 29 Dec 2004, Lee Ping Alison wrote:
> > >
> > >   > Dear Mr Day,
> > >   >
> > >   > While reading the Bioperl 1.4 documentation for the
> "Bio::AlignIO::maf" module, I found your email address and I have some
> questions about how to use "maf."
> > >   >
> > >   > Am I right to say that the strand information of each sequence in an
> "maf" file is not recorded, when the LocateableSeq object is created in the
> nextAln() method? I observed that $strand was not one of the arguments in
> the call to the constructor.
> > >   >
> > >   > If yes, what is the reason for not using the strand information? And
> subsequently, if I need to retrieve the strand information, how should I go
> about it?
> > >   >
> > >   > Thank you very much for answering my queries.
> > >   >
> > >   > Best Regards,
> > >   > Alison
> > >   > (Institute of Molecular and Cell Biology, Singapore)
> > >
> > > _______________________________________________
> > > Bioperl-l mailing list
> > > Bioperl-l at portal.open-bio.org
> > > http://portal.open-bio.org/mailman/listinfo/bioperl-l
> > >
> >
> >
> 
> _______________________________________________
> Bioperl-l mailing list
> Bioperl-l at portal.open-bio.org
> http://portal.open-bio.org/mailman/listinfo/bioperl-l
> 




More information about the Bioperl-l mailing list