Priorities

[will-moore]

Current status:

• Users can choose to load 1 OMERO.table and/or Key-Value pairs or Dataset Names for Images within a Project
• KVPs and Dataset names are copied to an MDV-project json file in "MDV" format. This file is read to provide data for datasources.json, binary column data and views.json - OMERO.tables column data is read on-the-fly from OMERO.tables.
• Views are saved back to the same MDV-project.json file. MDV projects can be listed and re-opened in MDV.

Priorities:

• Update MDV to latest version (needed for column text16 datatype). See https://github.com/will-moore/omero-mdv/pull/9
• Decide how to improve this process in the future, handle breaking API changes etc.
• Review the MDV-project.json format, decide on versioning etc.
• Understand how to combine ROIs and Images in MDV (multiple datasources, similar to https://mdv.molbiol.ox.ac.uk/projects/hyperion/6567?view=Spatial+stats+ROI ?)
• Add support for Tags and CSV files
• Support "Open with MDV..." for Datasets, Images, Screens and Plates
• Make OMERO-MDV pip installable, add to pypi.

Future ideas:

• Support combining multiple OMERO.tables
• Investigate editing of Annotations in MDV and saving back to KVP/Tags etc?
• Look into Viv viewer in MDV as a full viewer, with rendering settings, Z/T sliders etc, loading images as OME-Zarr.
• Support spatial data in OMERO and OMERO-MDV (e.g. hyperion example above)
• HCS Plate layout chart in MDV
• Support AnnData API for MDV to load data from OMERO and other static AnnData sources.

[xinaesthete]

Hi Will,

Input on things like versioning of JSON formats etc in MDV definitely could be useful.

In terms of Viv rendering settings, I'm hoping to use more of the Avivator code directly, pending that being exposed to be usable as a library... OME-Zarr support should be fairly trivial (started adding, but needs testing/cleanup/documentation etc).