[Bioperl-l] Re: Bio::FPC

Brian Osborne brian_osborne at cognia.com
Thu Aug 28 11:31:21 EDT 2003 

Jamie,

>From the looks of it seems that a Marker can have one or more Positions and
a Positions can have a range and a Map. But you asked about iterating over a
Marker's Positions to find the "clones" (and both "clones" and "contigs" are
smaller and larger Maps respectively, yes?). I'm seeing the ability to
get/set but not anything like "next_position", but this is after a minute's
inspection. I'd suggest you take a better and closer look at Mappable,
Postion, and Map to get the real answer. If there's really nothing
resembling this I'm certain you could create it, I think it's to your
advantage to use, and possibly enrich, Bioperl's existing objects so that in
the future you'll have access to all that's inside. But you knew that
already...

By the way, say "Hi!" to Rod Wing for me, he's a friend of mine from way
back.

Brian O.

-----Original Message-----
From: bioperl-l-bounces at portal.open-bio.org
[mailto:bioperl-l-bounces at portal.open-bio.org]On Behalf Of Jamie Hatfield
Sent: Thursday, August 28, 2003 10:22 AM
To: Brian Osborne
Cc: Ewan Birney; BioPerl-List
Subject: RE: [Bioperl-l] Re: Bio::FPC

Yes, that is valid.  So then I assume you're saying that the marker can
just hit the contig instead?  That would be fine if
1) the marker could hit a range
2) the marker could store all the clones it hits.

We need to be able to query the marker and ask it which clones and which
contigs it hits.  I don't think that would be possible with the Marker
Object, would it?

Thanks for keeping up this discussion, by the way.  I'd like to try to
get this finished up and added in correctly.  A few people have asked
how they get ahold of the fpc parser modules, and I've been telling them
to wait until I can get it part of bioperl in a proper manner.

Jamie

On Thu, 2003-08-28 at 06:39, Brian Osborne wrote:
> Jamie,
>
> Is it fair to say that "hitting a clone" is the same thing as "having a
> position in a clone that starts at the beginning of the clone and ends at
> the end of a clone"? Or "the FPCMarker's range goes from the beginning to
> the end of a clone"?
>
> Brian O.
>
> -----Original Message-----
> From: bioperl-l-bounces at portal.open-bio.org
> [mailto:bioperl-l-bounces at portal.open-bio.org]On Behalf Of Jamie Hatfield
> Sent: Wednesday, August 27, 2003 6:04 PM
> To: Ewan Birney
> Cc: Brian Osborne; BioPerl-List
> Subject: RE: [Bioperl-l] Re: Bio::FPC
>
> Bio::Map::Marker is really more like our clones.  A clone has a range
> that it exists in contig (or map).  But FPCMarkers don't have a position
> in a map.  They "hit" a clone.  That is why I felt it was necessary to
> create a new class.  I don't see how these two ideas overlap.
>
>
> On Tue, 2003-08-26 at 10:02, Ewan Birney wrote:
> >
> >
> > On Tue, 26 Aug 2003, Brian Osborne wrote:
> >
> > > Jamie,
> > >
> > > And a "marker" can be a genetic marker, yes? A la Bio::Map::Marker? If
> you
> > > take a look at this module you'll see that its definition of marker
> allows
> > > any marker to have different positions in different maps (contig
"map",
> > > genetic map, physical map). This seems to overlap with your notion of
> > > marker.
> > >
> > > Here's my first impression. There's a parser, MapIO::mapmaker for
> mapmaker,
> > > mapmaker makes maps from segregation data, genetic data. Your fpc
makes
> > > physical maps, yet physical and genetic maps can be merged to create
> > > "integrated maps". Your fpcmarker must be closely related to
> > > Bio::Map::Marker, in fact it's not clear that there should be an
> fpcmarker.
> > > I would think that a Marker object could be a reasonably rich one, and
> it
> > > could be created by fpc or any other program, it really shouldn't
matter
> > > much how it's created (in fact, all this new PopGen code must be
> ordering
> > > markers to make maps, I'd think). Perhaps you should be using some of
> the
> > > existing code in Bio/Map? Your thoughts?
> > >
> >
> > Brian - I doubt the pop gen stuff will overlap at all with this stuff.
but
> > the marker comment is right, though I can well believe there needs to be
> > specific FPC hooks for markers used for FPC stuff....
> >
> >
> >
> >
> > > Brian O.
> > >
> > > -----Original Message-----
> > > From: Jamie Hatfield [mailto:jamie at genome.arizona.edu]
> > > Sent: Tuesday, August 26, 2003 11:46 AM
> > > To: Brian Osborne
> > > Cc: BioPerl-List
> > > Subject: RE: [Bioperl-l] Re: Bio::FPC
> > >
> > > Yes, definitly, discussion is great!
> > >
> > > We had a little bit of a discussion about this back in November 2002,
> > > when I proposed the idea, and it was suggested by Heikki to try to fit
> > > it into either Bio::Map or Bio::Assembly.  Maybe I picked the wrong
> > > one?  How about this...  I will describe a little bit about what fpc
is,
> > > for those who don't know, and those who know Bio::Map and
Bio::Assembly
> > > will tell me if it fits in their design.  ok?
> > >
> > > FPC stands for FingerPrinted Contigs.  Its main purpose is to assemble
> > > clones into contiguous regions of overlaps, based on the fingerprint
of
> > > the clones.  These fingerprints can be from agarose (sp?) gels, or
HICF,
> > > or simulated, or whatever.  Maybe this is more like Assembly?
> > >
> > > Anyway, you have the clones, and there are also markers that hit the
> > > clones, and aid in assembling the clones into contigs.  These are the
> > > main 3 classes.  Clones, Contigs, Markers.  Contigs contain Clones.
> > > Markers hit Clones.  Clones are hit by markers.
> > >
> > > Contigs 1--m Clones
> > > Markers m--m Clones
> > >
> > > Is that a sufficient description of FPC, or do we need more to make a
> > > good decision?
> > >
> > > Thanks for initiating the discussion, Brian.
> > >
> > > Jamie
> > >
> > > On Tue, 2003-08-26 at 05:28, Brian Osborne wrote:
> > > > Jamie,
> > > >
> > > > One of the challenges in Bioperl is creating a single coherent set
of
> > > > modules from the many individual contributions. Could you tell us a
> bit
> > > > about your modules and how they overlap functionally with the
existing
> > > > modules in Bio::Map? If you take a look at those modules you can see
> that
> > > a
> > > > good number of the more steadfast Bioperl authors have contributed
to
> > > > Bio::Map, I'm sure that they'd like to see your modules integrate
> neatly
> > > > with the existing code.
> > > >
> > > > I'm not one of these authors, I'm simply responding because it seems
> that
> > > > you'd like to get some discussion going.
> > > >
> > > > Brian O.
> > > >
> > > > -----Original Message-----
> > > > From: bioperl-l-bounces at portal.open-bio.org
> > > > [mailto:bioperl-l-bounces at portal.open-bio.org]On Behalf Of Jamie
> Hatfield
> > > > Sent: Monday, August 25, 2003 6:35 PM
> > > > To: BioPerl-List
> > > > Subject: Re: [Bioperl-l] Re: Bio::FPC
> > > >
> > > > Again, how do I go about submitting this?
> > > >
> > > > On Thu, 2003-08-14 at 09:34, Jamie Hatfield wrote:
> > > > > Yes, actually.  We are just now finishing up the fpc parser.  I
was
> > > > > planning on soon asking the group how I would go about submitting
> it?
> > > > > It consists of 5 modules that we have put in the MapIO and Map
> > > > > namespaces.
> > > > > Bio::MapIO::fpc.pm
> > > > > Bio::Map::physical.pm
> > > > > Bio::Map::fpcmarker.pm  (sorry, but marker doesn't work)
> > > > > Bio::Map::clone.pm
> > > > > Bio::Map::contig.pm
> > > > >
> > > > > If you want to see how this object might be used, check out
> > > > > http://www.genome.arizona.edu/software/fpc/biofpc/index.html
> > > > >
> > > > > You'll see there documentation for the modules, and a few test
cases
> or
> > > > > example usages.
> > > > >
> > > > > Also, we are trying to make a generic converter to let you load in
a
> fpc
> > > > > file and generate the necessary GFF for GBrowse to display the fpc
> map.
> > > > > It's a quite simple display of the clones, markers, and contigs,
but
> > > > > maybe that will be usefull as an alternative to WebFPC (a java
view
> only
> > > > > version of fpc).  It works for us, but might not work for
everybody.
> We
> > > > > should be able to patch it up, though, if it's missing features.
> > > > >
> > > > > So, anyways, if somebody can let me know how to go about
submitting
> it,
> > > > > we'll start the process.  I looked through the FAQ and it
basically
> said
> > > > > to just post information if you have a module that you would like
to
> > > > > contribute, so, here's the information.
> > > > >
> > > > > Jamie
> > > >
> > > >
> > > > _______________________________________________
> > > > 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
> > >
> >
>
>
> _______________________________________________
> 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