Personal tools
  • We're Hiring!

You are here: Home Community Minutes Conference Calls 2006 2006-03-02 B,B,D,C

2006-03-02 B,B,D,C

Attending: Barry, Brian, Chris, Donald, Harry, Jason, Josh, Jean-Marie, Sheldon, Tony, Jeremy

Agenda

  • Review of last week's notes/issues
  • Week-in-review
    • Baltimore
    • Boston
      • Tony's data viewer
        • Old "feature" vs. new ROI5D - how does feature map to ROI5D?
        • Want to visualize features/ROI5D in image viewer - message bus event
        • How to classify images and ROI5Ds
        • Image grouping vs. classification
      • Sheldon's Shoola componentization
        • Developmental stability of code base - is it time to componentize Shoola, and would changes be integrated into trunk?
      • OMERO issues relating to Sheldon's external data analysis system
        • How to define external analysis modules? (e.g. CellProfiler)
        • How to add new semantic types from new modules?
        • How to import large amount of data from analysis (Spreadsheet importer?)
    • Dundee
      • Josh: "By gosh things are almost functional" (or, State of the Parsnip, most notably the query parsnips.)
      • Jason, Jean-Marie: Shoola2.5, Shoola3, and Shoola on SVN
      • Jason: State of SPW model
  • Week-to-come
  • And as always, WhoNeedsWhatFromWhomByWhen
    • Jason: Very critical to nail down dependencies and work for next few weeks
    • Josh: API requests: "going once, going twice,..." - See Api-Requests-Omero below


Notes

  • Harry's code is now in CVS at http://cvs.openmicroscopy.org.uk/viewcvs/?root=Harry
  • Sheldon/Tony have a number of new features that will hopefully be migrated them over to Omero.
    • Currently there is no "features type" in the new Omero (which may not be a requirement for Sheldon and Tony)
    • Once Jean-Marie is on his way, Josh should be able to get something set up to attach semantic types to RIO5Ds in a couple of weeks.
    • Integration and visualization of graphical primitives (circle, square, line, etc.) with Donald's Secret Viewer or the image viewer is ongoing
    • Tony's envisioned application would allow him to highlight points/coordinates/primatives and have then displayed on the image. This is basically what Donald is working on right now. After the current code merge this will become more of a priority for Donald, but right now, its a 'side project'.
    • Sheldon is wondering on the status of componentization (i.e.: using the Eclipse plug-ins model for the viewer). We are currently getting rid of OMEJava and cleaning the rendering engine. After that the code should be more stable so components can be made. Sheldon may be able to start on things before that. For example, the Environment package is staying but some of the data types will be going away - so only certain things can be done for now.
    • Sheldon pointed out a need to import large amounts of data at one time (as opposed to the small data set imports that occur now)
  • Things in Dundee are starting to wrap up for our demo.
    • We can now store types and do "all sorts of fun things with them".
    • There are still a number of issues with the OME 2/3 semantic types, but most of milestone 1 issues are completed.
    • Chris has been working with Resurrect, and has all Image/Pixels types along with Rendering Engine types included
    • Chris and Jean-Marie have got all data model changes to Josh and they should be represented in the database
    • Jean-Marie is working on the new version of Shoola.
    • State of the SPEW: Jason has talked to a number of screening folks and has paired down the requirements and has sent to Harry. There is another conference scheduled to try and finalize requirements this week. Stay Tuned!

Action Items

  • Have Tony, Donald, and Jean-Marie touch base on the technical details of how the primative/point display will overlay on images.
  • Brian / Chris should do up a Tiki page on Rebirth (what does rebirth do/not do, what is rebirth)
  • Jason / Chris are going to have a conversation in the next few days about how to implement a better data import feature (one for OMEXML is definitely needed). This will hopefully meet Sheldon's needs.
  • Jason needs to organize massage-lady
  • Jason is going to send out an email to update everyone on the status of 3rd party database integration



Api-Requests-Omero (for discussion above)


Stuff that Boston uses:

In order to define the API that Boston would like to see in OMERO, we have decided to list all the calls that we use in Shoola/OME-JAVA.

Note: JM Access to OMERO is done via the OMEROGateway. Shoola's container initializes Services that Agents can access. Some of the calls needed are already implemented in the Omero Service I/F (cf. env.data package).
Need to update the doc b/c few things have changed between OMERO 2 and OMERO 3.


  • DatasetSummary and ImageSummary passed into DataViewer through event bus when its loaded.
    • iQuery.getById( Dataset.class, id );
  • DataManagementService.retrieveImages(datasetSummary.getID())
    • To get a list of images in a dataset: (OME-JAVA)
    • iPojos.retrieveCollection ( dataset, Dataset.IMAGES, null );
  • DataManagementService().retrieveImage(imageID)
    • used only to get default pixelsID for use when sending an event to the EventBus for Image Viewer to load on the image that is being viewed in DataViewer
    • ( (Image) iQuery.getById( Image.class, imageID ) ).getActivePixels().getId()
  • PixelService.getComposite()
    • // Coming...
    • renderingEngine.render( new PlaneDef() );
  • ModuleRetrievalManager to get currently installed modules
    • iQuery.getByClass( EventType.class );
  • OME-JAVA code to add modules and semantic types.
    • iUpdate.saveObject( new EventType() ); // Gets you a new module
    • // semantic types have to be defined on the server-side
  • Spreadsheet importer to import data
    • ToBeDiscussed

  • SemanticTypesService().getAvailableImageTypes()
  • SemanticTypesService().getAvailableFeatureTypes()
    • iTypes.getResultTypes ( ); // Result types get applied to Features and Images at the same time
    • iTypes.getAnnotationTypes ( );
    • iTypes.getContainerTypes ( );
    • iTypes.getPojoTypes ( );
    • iTypes.getImportTypes ( );
    • // etc...
    • // for this will need a types.vm

  • SemanticTypesService().countImageAttributes(semanticType,imageIDList)
  • SemanticTypesService().countFeatureAttributes(semanticType,featureIDList)
    • ResultsService.countResults( SomeSpecificResult.class, imageSetIdList );

  • SemanticTypesService().retrieveFeatureAttributes(semanticType, featureIDList)
  • SemanticTypesService().retrieveImageAttributes(semanticType, imageIDList)
    • ResultsService.retrieveResults( Result.class, imageSetIdList );

  • DataFactory().retrieveList(Feature.class, c);
    • To get a list of features in an image: (would like a similar call, instead use Criteria)
    • // RoiService.retrieveImageRois( int imageId )
    • iQuery.getByFieldEq( Roi5D.class, Roi5D.PIXELS, imageId )
    • iQuery.getByExample( example );
    • iQuery.queryList( " select r from Roi5 wheren r.pixels.id = ? ", pixelsId );
    • // also
    • Roi5D example = new Roi5D();
    • Pixels proxy = new Pixels( pixelID );
    • proxy.unload();
    • example.setPixels( proxy );

  • Extra calls that will be needed (Josh)
    • ResultService.findImageSet( imageIdList, roi5dIdList, boolean and )

Note: possible names {ResultService, IResults, FeatureService, AnalysisService, IAnalysis, ... }
Document Actions