Different rendering for wiggle tracks
|
View:
New views
7 Messages
—
Rating Filter:
Alert me
|
 |
Different rendering for wiggle tracks
by Stephen Taylor-5
::
Rate this Message:
Hi,
We have two installations of gbrowse for our production and test environment. (1) The gbrowse production server was from the 1.69 tar ball using ActivePerl-5.8.8.822, and run with bioperl-live as on 2008-09-04 (2) On the gbrowse development server we installed GBrowse-CVS as on 2009-03-19 using netinstall to see if this would fix some problems However, on (2) we noticed there is a difference in the way wiggle tracks are rendering. The (2) version looks smoothed and wrong compared with the raw data. Was this type of rendering desired? We notice the wiggle_xyplot.pm code is completely different. To see what I mean have a look at: (1) http://sara.molbiol.ox.ac.uk/userweb/staylor/gbrowse/gbrowse.jpg (2) http://sara.molbiol.ox.ac.uk/userweb/staylor/gbrowse/gbrowsedev.jpg The age old question :-) do we wait for 1.7 to fix this or is this type of rendering likely to stay? Kind regards and thanks, Steve ------------------------------------------------------------------ Medical Sciences Division Weatherall Institute of Molecular Medicine/Sir William Dunn School Oxford University ------------------------------------------------------------------------------ Register Now for Creativity and Technology (CaT), June 3rd, NYC. CaT is a gathering of tech-side developers & brand creativity professionals. Meet the minds behind Google Creative Lab, Visual Complexity, Processing, & iPhoneDevCamp as they present alongside digital heavyweights like Barbarian Group, R/GA, & Big Spaceship. http://p.sf.net/sfu/creativitycat-com _______________________________________________ Gmod-gbrowse mailing list Gmod-gbrowse@... https://lists.sourceforge.net/lists/listinfo/gmod-gbrowse |
|
|
Re: Different rendering for wiggle tracks
by Jason Stajich-3
::
Rate this Message:
Steve -
I have had the same problems - I thought some of it related to changes in the Bio::Graphics::Wiggle::Loader and how trimming was being done. There are defaults for the trimming: my $trim = $self->current_track->{display_options}{trim} || 'stdev2'; You can try and specify this in the your wigfile header trim=none or other combinations. I can't remember what was the best setting to go back to the original way or if the interaction with the wiggle_xyplot changes are also the major change. -jason On May 28, 2009, at 8:06 AM, Steve Taylor wrote:
------------------------------------------------------------------------------ Register Now for Creativity and Technology (CaT), June 3rd, NYC. CaT is a gathering of tech-side developers & brand creativity professionals. Meet the minds behind Google Creative Lab, Visual Complexity, Processing, & iPhoneDevCamp as they present alongside digital heavyweights like Barbarian Group, R/GA, & Big Spaceship. http://p.sf.net/sfu/creativitycat-com _______________________________________________ Gmod-gbrowse mailing list Gmod-gbrowse@... https://lists.sourceforge.net/lists/listinfo/gmod-gbrowse |
|
|
Re: Different rendering for wiggle tracks
by Lincoln Stein-3
::
Rate this Message:
Hi Steve,
The wiggle code was recently moved out of GBrowse and into a separate Bio::Graphics distribution. At the same time we removed the default smoothing to better render the raw data. Could you try to download the CPAN version of Bio::Graphics and install it on top of the GBrowse-CVS version? If the data continues to look wrong, I will fix it. Lincoln On Thu, May 28, 2009 at 11:06 AM, Steve Taylor <stephen.taylor@...> wrote: Hi, -- Lincoln D. Stein Director, Informatics and Biocomputing Platform Ontario Institute for Cancer Research 101 College St., Suite 800 Toronto, ON, Canada M5G0A3 416 673-8514 Assistant: Renata Musa <Renata.Musa@...> ------------------------------------------------------------------------------ Register Now for Creativity and Technology (CaT), June 3rd, NYC. CaT is a gathering of tech-side developers & brand creativity professionals. Meet the minds behind Google Creative Lab, Visual Complexity, Processing, & iPhoneDevCamp as they present alongside digital heavyweights like Barbarian Group, R/GA, & Big Spaceship. http://p.sf.net/sfu/creativitycat-com _______________________________________________ Gmod-gbrowse mailing list Gmod-gbrowse@... https://lists.sourceforge.net/lists/listinfo/gmod-gbrowse |
|
|
Re: Different rendering for wiggle tracks
by Stephen Taylor-5
::
Rate this Message:
Hi Lincoln/Jason,
> > The wiggle code was recently moved out of GBrowse and into a separate > Bio::Graphics distribution. At the same time we removed the default > smoothing to better render the raw data. Could you try to download the > CPAN version of Bio::Graphics and install it on *top* of the GBrowse-CVS > version? If the data continues to look wrong, I will fix it. > We tried it with Bio::Graphics 1.94 and no difference...at least when I just reloaded the page... Jason, the code you pointed out my $trim = $self->current_track->{display_options}{trim} || 'stdev2'; seems to fly in the face of the docs http://gmod.org/wiki/Gbrowse/Uploading_Wiggle_Tracks#Wiggle_Track_Options_Supported_by_GBrowse trim stdev|stdev2|stdevn # default is none So would it be better to change it to my $trim = $self->current_track->{display_options}{trim} || 'none'; ? What I am not 100% clear on is do we have regenerate all the wig binaries to see an effect or just reload the page after these fixes (assuming mod_perl is off!)? Thanks, Steve > Lincoln > > On Thu, May 28, 2009 at 11:06 AM, Steve Taylor > <stephen.taylor@... <mailto:stephen.taylor@...>> wrote: > > Hi, > > We have two installations of gbrowse for our production and test > environment. > > (1) The gbrowse production server was from the 1.69 tar ball using > ActivePerl-5.8.8.822, and run with bioperl-live as on 2008-09-04 > (2) On the gbrowse development server we installed GBrowse-CVS as on > 2009-03-19 using netinstall to see if this would fix some problems > > However, on (2) we noticed there is a difference in the way wiggle > tracks are rendering. The (2) version looks smoothed and wrong > compared with the raw data. > Was this type of rendering desired? We notice the wiggle_xyplot.pm > <http://wiggle_xyplot.pm> code is completely different. > > To see what I mean have a look at: > > (1) http://sara.molbiol.ox.ac.uk/userweb/staylor/gbrowse/gbrowse.jpg > (2) http://sara.molbiol.ox.ac.uk/userweb/staylor/gbrowse/gbrowsedev.jpg > > The age old question :-) do we wait for 1.7 to fix this or is this > type of rendering likely to stay? > > Kind regards and thanks, > > Steve > ------------------------------------------------------------------ > Medical Sciences Division > Weatherall Institute of Molecular Medicine/Sir William Dunn School > Oxford University > > ------------------------------------------------------------------------------ > Register Now for Creativity and Technology (CaT), June 3rd, NYC. CaT > is a gathering of tech-side developers & brand creativity > professionals. Meet > the minds behind Google Creative Lab, Visual Complexity, Processing, & > iPhoneDevCamp as they present alongside digital heavyweights like > Barbarian > Group, R/GA, & Big Spaceship. http://p.sf.net/sfu/creativitycat-com > _______________________________________________ > Gmod-gbrowse mailing list > Gmod-gbrowse@... > <mailto:Gmod-gbrowse@...> > https://lists.sourceforge.net/lists/listinfo/gmod-gbrowse > > > > > -- > Lincoln D. Stein > Director, Informatics and Biocomputing Platform > Ontario Institute for Cancer Research > 101 College St., Suite 800 > Toronto, ON, Canada M5G0A3 > 416 673-8514 > Assistant: Renata Musa <Renata.Musa@... > <mailto:Renata.Musa@...>> > > > ------------------------------------------------------------------------ > > ------------------------------------------------------------------------------ > Register Now for Creativity and Technology (CaT), June 3rd, NYC. CaT > is a gathering of tech-side developers & brand creativity professionals. Meet > the minds behind Google Creative Lab, Visual Complexity, Processing, & > iPhoneDevCamp as they present alongside digital heavyweights like Barbarian > Group, R/GA, & Big Spaceship. http://p.sf.net/sfu/creativitycat-com > > > ------------------------------------------------------------------------ > > _______________________________________________ > Gmod-gbrowse mailing list > Gmod-gbrowse@... > https://lists.sourceforge.net/lists/listinfo/gmod-gbrowse ------------------------------------------------------------------------------ Register Now for Creativity and Technology (CaT), June 3rd, NYC. CaT is a gathering of tech-side developers & brand creativity professionals. Meet the minds behind Google Creative Lab, Visual Complexity, Processing, & iPhoneDevCamp as they present alongside digital heavyweights like Barbarian Group, R/GA, & Big Spaceship. http://p.sf.net/sfu/creativitycat-com _______________________________________________ Gmod-gbrowse mailing list Gmod-gbrowse@... https://lists.sourceforge.net/lists/listinfo/gmod-gbrowse |
|
|
Re: Different rendering for wiggle tracks
by Lincoln Stein-3
::
Rate this Message:
Hi Steve,
Sadly the default "stdev2" default is still in the Bio::Graphics distribution! I swear I removed it, but perhaps I failed to commit the code. I am working on another bug in Bio::Graphics and will submit a new version to CPAN. Unfortunately the stdev2 filtering occurs at LOAD time, which means you have to reload all wiggle files. Making stdev2 the default was a bad mistake and I regret the inconvenience. Lincoln
On Fri, May 29, 2009 at 11:01 AM, Steve Taylor <stephen.taylor@...> wrote: Hi Lincoln/Jason, -- Lincoln D. Stein Director, Informatics and Biocomputing Platform Ontario Institute for Cancer Research 101 College St., Suite 800 Toronto, ON, Canada M5G0A3 416 673-8514 Assistant: Renata Musa <Renata.Musa@...> ------------------------------------------------------------------------------ Register Now for Creativity and Technology (CaT), June 3rd, NYC. CaT is a gathering of tech-side developers & brand creativity professionals. Meet the minds behind Google Creative Lab, Visual Complexity, Processing, & iPhoneDevCamp as they present alongside digital heavyweights like Barbarian Group, R/GA, & Big Spaceship. http://p.sf.net/sfu/creativitycat-com _______________________________________________ Gmod-gbrowse mailing list Gmod-gbrowse@... https://lists.sourceforge.net/lists/listinfo/gmod-gbrowse |
|
|
Re: Different rendering for wiggle tracks
by Jason Stajich-3
::
Rate this Message:
ah hah! I knew that change was wrong, good to get confirmation.
-jason On May 29, 2009, at 11:26 AM, Lincoln Stein wrote:
------------------------------------------------------------------------------ Register Now for Creativity and Technology (CaT), June 3rd, NYC. CaT is a gathering of tech-side developers & brand creativity professionals. Meet the minds behind Google Creative Lab, Visual Complexity, Processing, & iPhoneDevCamp as they present alongside digital heavyweights like Barbarian Group, R/GA, & Big Spaceship. http://p.sf.net/sfu/creativitycat-com _______________________________________________ Gmod-gbrowse mailing list Gmod-gbrowse@... https://lists.sourceforge.net/lists/listinfo/gmod-gbrowse |
|
|
Re: Different rendering for wiggle tracks
by Lincoln Stein-3
::
Rate this Message:
Hi Jason,
Actually, I take it all back. The badness is not the stdev2 filtering, but the log transform operation, which was on by default in older versions of Bio::Graphics. I think the stdev2 clipping is still the right thing to do by default, because otherwise occasional outliers cause the internal scaling to be thrown off. The indexed version of wig files (wib) only have a dynamic range of 1..255 because they store values in a single byte. Unless the scaling is correct, an outlier will force most of the data to occupy a tiny slice of that range. Lincoln On Fri, May 29, 2009 at 2:30 PM, Jason Stajich <jason@...> wrote:
-- Lincoln D. Stein Director, Informatics and Biocomputing Platform Ontario Institute for Cancer Research 101 College St., Suite 800 Toronto, ON, Canada M5G0A3 416 673-8514 Assistant: Renata Musa <Renata.Musa@...> ------------------------------------------------------------------------------ Register Now for Creativity and Technology (CaT), June 3rd, NYC. CaT is a gathering of tech-side developers & brand creativity professionals. Meet the minds behind Google Creative Lab, Visual Complexity, Processing, & iPhoneDevCamp as they present alongside digital heavyweights like Barbarian Group, R/GA, & Big Spaceship. http://p.sf.net/sfu/creativitycat-com _______________________________________________ Gmod-gbrowse mailing list Gmod-gbrowse@... https://lists.sourceforge.net/lists/listinfo/gmod-gbrowse |
| Free embeddable forum powered by Nabble | Forum Help |
