2021-01-12
Attending
- Petr, Josh (notes), Frances, June, Simon, Dom, Jason, J-m, Melissa, David, Wilma, Chris, Emil, Erin, Will, Kevin
Start: 2:00 pm UK
Project Status
(2-3 minutes each)
- IDR
- Working on next release (ETA: end of January)
- NGFF
- Downsampling in Z:
- https://github.com/ome/ome-zarr-py/pull/71
- Looking maybe into bdv
- https://javadoc.scijava.org/Fiji/index.html?bdv/export/Downsample.html
- Leiden & “Federation”
- various pieces of the puzzle, and we need to figure out where we’re investing (oldest to newest)
- OmeroReader
- exporter (OmeroMetadata)
- downloader
- duplicate
- omero-cli-zarr
- Deep zoom
- Simon: isn’t this what Will did with HCS?
- JRS: there’s a difference
- Chris: disconnect between the preview & the actual data. More than an affine transform. Calculating approximation with computer vision
- Simon: working towards multimodal viewing
- JRS: multimodal may be the best way to express it. in digpath, lowres+hires. Alternatively, can place the macro somewhere in the display
- Josh: with a context box like bird?
- Chris: file format specific. optimal to have them.
- Chris: also a question of if you’ve captured multiple images. (i.e. lots of things)
- Emil: number of customers in digpath where it’s already an issue. Multiple tisse samples from the same slide. Very difficult to know which is which in OMERO. Linkage is important.
- Erin: agreed. Cute to have the overview but isn’t the necessary piece. The mapping is more important. Annotations are essential, especially if it should be a single “canvas”
- Emil: there is software that treats it like OMERO, other does not.
- Simon: sounds like ROIs
- Josh: to some extent, ROIS could just be images.
- Petr: yes, image image link
- Releases
- Simon: discussing omero-py release (5.9). Objections?
- J-m: need secrets to be set up for conda packages. will synchronize with Seb.
- SA
- Flash is dead (utterly)
- i.e. don’t tell anyone to upload to QA
- David: alternative?
- Josh: zenodo?
- JRS: downside? Private files.
- Chris: i.e. publicly available.
- Simon: see upload_test
- Glencoe
- Updating bioformats2raw with jzarr.
- Needs performance testing.
- Using that from microservices (e.g. during rendering)
- Using the opportunity to align with ome-zarr layout.
- JRS: Glencoe contacted by imaging data commons (NCI/IDC)
- Announced during public presentation 1-2 months ago
- adding readers/writers for variants of DICOM-WSI
- specifying the work now, but settling down into a definition
- Q1-2/2021
- Community (14:28)
- LIM/Nikon (David)
- company producing OME-TIFF files that need validating (incl. FileAnnotations with proprietary byte arrays).
- Tested via import to OMERO and re-export. Import, visualization, and export all need work. Prepared to undertake that?
- Pondering for a bit before deciding, may need to discuss the overall goal.
- Open to better suggestions
- Workshops
- GBI thread (27-29.01)
- Stockholm (10.02)
- McGill (18-19.02)
- Gothenburg (TBD) - likely to want late Jan/Feb.
AOB
(5 min. max; tech. Discussion should be highlighted to relevant people and rescheduled)
Main Topic
(20-25 minutes plus 15 minutes questions max)
2021 Roadmap
- JRS: objections for these high-level activities for the next few weeks
- Missing: training (Added)
- Workflows, J-m? Added under various sections
- Need to decide where we’re going to run them.
- Conversation about what will be exposed as Zarr work.
- NB: Ice 3.7 may be a problem. S3 could be an alternate route, simplifying the data access.
- Notebooks are a cross-cutting concern.
- Simon: docs reoganization?
- J-m: going back to it now that GitHub Actions are done.
- JRS: feedback by next week then. :+1:
