Agenda (from Curtis' email, 01/31/2006 10:05 PM)

1. Support for uploading image data to the OME database....
2. Support for accessing image data (pixels and metadata) from an OME database, including searching on various criteria ...
3. Eventual support for more complex interactions such as iterative editing of overlays by different users...

Overview of OME from Josh

• Ok introduction on the Tiki webpage (http://cvs.openmicroscopy.org.uk/tiki/tiki-index.php?page=Omero+Design). Work in progress; more to come, etc.
• There is currently a discrepancy between the OMERO database itself and the XML data model. The goal here is to eventually have a direct mapping between the XML data model and the running database.
• There is some concern that the OME XML spec doesn't change. We haven't put in place a regular OME spec update process and need a curator/main repository/community for where the spec is held and maintained.
• omero is not OME-XML and not changing it; just as ome is not OME-XML

Support for uploading image data to the OME database.

• Examine using Bio-Formats package as a part of import
• Critical to see if they support the minimum metadata requirements

Support for accessing image data (pixels and metadata) from an OME database

• This is the goal of the Dundee team (should start seeing stuff mid-march)
• We are hoping that any queries are handled on the server side for performance
• We need more use case examples for queries
  • Queries: centered on images
    • dates, user, acquired date, own annotations (missing from web client)

Eventual support for more complex interactions such as iterative editing of overlays by different users.

• Further discussion needed on this topic so we can get some solid ideas on what VisBio is doing and how we can store that information. (This discussion is planned for an upcoming meeting).
• Possible to store much of the information on the server as Roi5Ds and Omero Annotations.
• visbio package for overlays: loci.visbio.overlays

OME XML Import in Omero

• Current status: future. (Also good to get DataModelProposal changes into the spec first)
• For Visbio/Bio-Formats it means Flexibility
• For server-side, it should mean just implementing one (more?) form of importer

What are the minimum set of requirements to get VisBio integration?

• Examine XML
• Omero XML Importer
• Visbio imports and tests Query API

Action Items

• Get Kevin and Donald together to talk about glyphs/overlays.
  • Curtis: I have sent an email with some notes about VisBio's overlays:
    1. You can read about VisBio's overlay support in its built-in help. Just launch the program, hit F1, and check out the "Overlays" topic in the "Data transforms" section.
    2. Right now VisBio only supports overlays in 2D. Eventually we want to provide the ability to link 2D structures together to create higher dimensional structures.
    3. The format VisBio uses to store overlays on disk is actually a really simple tab-delineated text format suitable for import into a spreadsheet application.
    4. The code for the overlay support is in loci/visbio/overlays in the VisBio source.
• Get Brian/Chris and VisBio "importer guy" talking
  • Curtis: Melissa Linkert is the person currently working on LOCI's Java Bio-Formats package. Chris & I are ironing out the API and other details for the importer logic, and I will be CCing Melissa on future correspondence.
• Kevin is going to send Josh some links/information on the "annotation expert" that is working with VisBio [DONE]
  • Curtis: His name is Badri Roysam, and the project is called FARSIGHT.
• Curtis is going to send Brian/Chris a link for their current import code [DONE]
• Kevin and Jason get together to talk about "What OME XML is"
• Curtis provides Josh with example XML [DONE]
• Talk more often! Either separate meetings infrequently, or Madison joins weeklys.
  • Next Meeting: 24th of Feb. 1600 CET Friday.

Tabled

• Discussion on changes to XML spec and the update process.