[ome-users] OME-XML Schema questions

Tue Feb 26 19:16:09 GMT 2008

Hi Pierre,

I am CCing the ome-users list to involve more of the OME developers in the
discussion.

About an element LaserCombiner:
>
> ·         The main problem is that the wavelength and Attenuation of the
> light source are defined in the LightSourceRef element. If I have only one
> LightSource element (the LaserCombiner), I can't indicate the intensities of
> all lasers for a given LogicalChannel. The standard scenario with the laser
> combiner is to make a time lapse with a switch between 2 intensities
> configurations.
>
> ·         Our clients also have the possibility to control multiple LEDs
> what would define in this case a "LEDCombiner". A LightCombiner would also
> be better.
>
>
Good points. I agree with the more general name LightCombiner. My earlier
proposal would have a LightCombiner made up of LightSources via
LightSourceRefs, meaning a given LightCombiner would have fixed wavelength
and attenuation for each component light source. Is that realistic? Or do
you typically adjust the wavelength and attenuation of each component light
source within a defined combiner module?

If you do adjust them, we would need a mechanism for enumerating the
settings for each component light source for a given LogicalChannel. The
easiest way I can think of to model that at the moment would be to have a
LightCombinerRef element with LightSourceRef children identifying the
component light sources and their settings. The ID of the LightCombinerRef
would reference a LightCombiner (part of your Instrument), but the
LightCombiner itself would not necessarily need to identify its components,
since they are enumerated in each Ref. Or it could enumerate them there too
using LightSourceRefs without any particular wavelength or attenuation
values. Kind of ugly, but it gets the job done.

 About Tile:
>
> ·         Each acquired picture of a scanned region have different times
> but the T dimension is reserved to indicate a loop on the scan (the region
> is scanned several times thus giving information about the evolution of the
> sample on a large region over the time).
>
>
What we really need is explicit support for tiled Images. It is something
that would be good to support anyway, since some third party formats use it,
and ideally Bio-Formats would properly represent tiled images within the OME
data model. Anyone else on the list want to propose such an extension to the
XML schema?

 A last question: will my pictures still be loaded on OMERO if I have a
> number of ChannelComponent different from SizeC (I understand that this
> cannot be interpreted but are the additional ChannelComponent ignored or
> does it produce an error) ?
>
>
I have no idea; perhaps one of the OMERO developers can comment. I would
suggest trying it yourself, but unless you want to write some Java code, it
would be easiest to wait until import of OME-XML and OME-TIFF are supported
within OMERO (in Beta3, due this summer).

 -Curtis

On Fri, Feb 22, 2008 at 11:48 AM, Pierre Menard <menard at till-photonics.com>
wrote:

>  Thank you very much for your answer.
>
>
>
> I just wanted to clear some points:
>
> About an element LaserCombiner:
>
> ·         The main problem is that the wavelength and Attenuation of the
> light source are defined in the LightSourceRef element. If I have only one
> LightSource element (the LaserCombiner), I can't indicate the intensities of
> all lasers for a given LogicalChannel. The standard scenario with the laser
> combiner is to make a time lapse with a switch between 2 intensities
> configurations.
>
> ·         Our clients also have the possibility to control multiple LEDs
> what would define in this case a "LEDCombiner". A LightCombiner would also
> be better.
>
>
>
> About Tile:
>
> ·         Each acquired picture of a scanned region have different times
> but the T dimension is reserved to indicate a loop on the scan (the region
> is scanned several times thus giving information about the evolution of the
> sample on a large region over the time).
>
>
>
> A last question: will my pictures still be loaded on OMERO if I have a
> number of ChannelComponent different from SizeC (I understand that this
> cannot be interpreted but are the additional ChannelComponent ignored or
> does it produce an error) ?
>
>
>
> Thank you again for your help
>
>
>
> Best Regards
>
>
>
> Pierre
>
>
>
>
>
> -----------------------------------------
>
> Pierre MENARD
>
> Software Entwickler
>
> TILL Photonics GmbH
>
> Lochhamer Shlag 19
>
> D - 82166 Gräfelfing
>
> menard at till-photonics.com
>
> www.till-photonics.com
>
>
>
> CEO: Dr. Rainer Uhl
>
> Amtsgericht München
>
> HRB-No. 104383
>
> Ust.-IdNo. DE 811580073
>
> WEEE-Reg.-No DE 39138453
>
> -----------------------------------------
>
>
>
> *From:* ctrueden.wisc at gmail.com [mailto:ctrueden.wisc at gmail.com] *On
> Behalf Of *Curtis Rueden
> *Sent:* Freitag, 22. Februar 2008 17:54
> *To:* Pierre Menard
> *Cc:* Andrew Patterson; Kevin Eliceiri; Jason Swedlow
> *Subject:* Re: OME-XML Schema questions
>
>
>
> Hi Pierre,
>
>           Multiple LightSourceRef for a LogicalChannel
>
> Most of our clients use a laser combiner to acquire TIRF pictures meaning
> a channel is defined by several light sources each with a given intensity.
> Is there a possibility to associate several LightSourceRef to a single
> LogicalChannel ?
>
>      With a laser combiner, there are of course several channels (for the
> light source) but the CCD camera is only acquiring one picture thus one
> channel in my Pixel array (SizeC = 1).
>
> Or should I define a LogicalChannel for each laser, each one containing a
> ChannelComponent pointing to the same channel dimension of the Pixel array
> (meaning there can be several LogicalChannel for the same channel dimension
> of the Pixel array) ?
>
>  There would not be multiple ChannelComponents -- they are typically used
> for the opposite of your scenario: one detector (e.g., a color camera)
> producing multiple ultimate intensity values (e.g., red, green and blue).
> The general rule is that the number of ChannelComponents must equal SizeC.
> Since your SizeC is 1, you have a single LogicalChannel with one
> ChannelComponent.
>
> So we are back to the original problem of representing your laser
> combiner's component light sources. To do it properly, we should probably
> add a new kind of LightSource called LaserCombiner that allows you to list
> light sources that define it. Then LogicalChannel continues to need only a
> single LightSource reference (to the LaserCombiner), but you can still
> express exactly which light sources are involved with the channel.
>
> I have CCed Kevin Eliceiri, who knows more about optics and microscopy
> techniques. Kevin and Andrew, any comments?
>
>    ·         Polygon ROI
>
> Our software can perform FRAP on polygons. Will it be possible in OME-XML
> to define polygon ROIs (list of points) ?
>
>    It will be possible, but there is no standard yet. We have had some
> thorough discussions on the best way to add more complex ROIs
> (non-rectangular, multidimensional, potentially disjoint, etc.) to the
> schema, but do not yet have a solution. It is something we want to add soon,
> as both OMERO and VisBio now both have similar manual annotation tools that
> could make use of it.
>
>    ·         Tiled pictures
>
> Is it possible in the OME schema to have several pictures in a single ZCT
> plane. It is common for our clients to scan a XY region, thus acquiring
> several pictures with the same Z, the same C and the same T information.
> These pictures are then stitched in a single picture but we would like to
> conserve the original pictures. The only way we found in OME is to define a
> Pixels element for each XY position but we then lose the logic of the data.
>
>    As of the 2007-06 schema version, you can specify a custom stage
> position for each plane in the pixels set. When you say the pictures have
> "the same T information," is that really true, since it takes a non-zero
> amount of time to scan the region? If not, one way is to have a separate
> time point for each XY position. But then it would be up to the client to
> reconstruct the images properly. Andrew, do you know a better way?
>
> Another possibility could be to store each composite image plane as a
> single image plane in the OME-TIFF file, but use TIFF's tiling capability to
> store each piece of the image plane as its own tile. (Are the tiles all the
> same size?) But then you are not annotating in the metadata why the image
> planes are tiled like that, unless you introduce a custom semantic type for
> it somewhere.
>
> -Curtis
>
> On Tue, Feb 19, 2008 at 4:56 PM, Jason Swedlow <jason at lifesci.dundee.ac.uk>
> wrote:
>
> Hi Pierre-
>
>
>
> Thanks for your email.
>
>
>
> I am cc'ing your email to Andrew Patterson, who covers OME-XML and Curtis
> Rueden, who covers OME-TIFF.
>
>
>
> I am pretty sure that most of what you want is already in there, but they
> are the experts and can confirm.  See for example:
>
>
>
> http://www.openmicroscopy.org/Schemas/OME/2007-06/ome.xsd
>
>
>
> Documentation is available:
>
>
>
> http://www.openmicroscopy.org/Schemas/
>
>
>
> I am confused on the first issue-- I can see the use of a laser combiner,
> but is a single channel coming from multiple lasers?  Maybe I missed your
> point.
>
>
>
> Cheers,
>
>
>
> Jason
>
>
>
>
>
> On 19 Feb 2008, at 15:09, Pierre Menard wrote:
>
>
>
>  Hi,
>
>
>
> We are currently developing a new software to control our microscopy
> environment and we decided to use OME-XML and OME-TIFF as default metadata
> format. We still have problem to integrate some of our data. Could you
> please inform us if a changes are planned for the following points or if a
> solution already exists ?
>
>
>
> ·         Multiple LightSourceRef for a LogicalChannel
>
> Most of our clients use a laser combiner to acquire TIRF pictures meaning
> a channel is defined by several light sources each with a given intensity.
> Is there a possibility to associate several LightSourceRef to a single
> LogicalChannel ?
>
>
>
> ·         Polygon ROI
>
> Our software can perform FRAP on polygons. Will it be possible in OME-XML
> to define polygon ROIs (list of points) ?
>
>
>
> ·         Tiled pictures
>
> Is it possible in the OME schema to have several pictures in a single ZCT
> plane. It is common for our clients to scan a XY region, thus acquiring
> several pictures with the same Z, the same C and the same T information.
> These pictures are then stitched in a single picture but we would like to
> conserve the original pictures. The only way we found in OME is to define a
> Pixels element for each XY position but we then lose the logic of the data.
>
>
>
> Thank you for your help
>
>
>
> Best Regards
>
>
>
>
>
> -----------------------------------------
>
> Pierre MENARD
>
> Software Entwickler
>
> TILL Photonics GmbH
>
> Lochhamer Shlag 19
>
> D - 82166 Gräfelfing
>
> menard at till-photonics.com
>
> www.till-photonics.com
>
>
>
> CEO: Dr. Rainer Uhl
>
> Amtsgericht München
>
> HRB-No. 104383
>
> Ust.-IdNo. DE 811580073
>
> WEEE-Reg.-No DE 39138453
>
> -----------------------------------------
>
>
>
>
>
>
>
>
>
> **************************
>
> Wellcome Trust Centre for Gene Regulation & Expression
>
> College of Life Sciences
>
> MSI/WTB/JBC Complex
>
> University of Dundee
>
> Dow Street
>
> Dundee  DD1 5EH
>
> United Kingdom
>
>
>
> phone (01382) 385819
>
> Intl phone:  44 1382 385819
>
> FAX   (01382) 388072
>
> email: jason at lifesci.dundee.ac.uk
>
>
>
> Lab Page: http://www.dundee.ac.uk/lifesciences/swedlow/
>
> Open Microscopy Environment: http://openmicroscopy.org
>
> **************************
>
>
>
>
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openmicroscopy.org.uk/pipermail/ome-users/attachments/20080226/b5275436/attachment.html>