2015-10-13 Tuesday Team Meeting
Dundee: Kenny, Gus, Simon, June, Will, Jean-Marie, Graeme, Balaji, Dominik, Petr, Roger, Colin, Jason, Simone, Helen, Remote: Josh M, David G, Kelli G, Andreas K, Emil R., Melissa L., Eleanor W., Ian M., Chris A., Curtis R, Ola T. (14:41)
Agenda - 2:30pm Start
Accepting minutes from last meeting
Project Timelines (2-3 minutes each)
OMERO
- 5.2.0 (J-m) Django upgrade from 1.6 to 1.8. Testing on-going Admins will install version of choice How long with Ice 3.5 be supported? Roger: they say they are still supporting it Chris: could change New PRs for web and Insight for right panel. PR deadline by next Monday in time for Nov 2nd release For 5.2.x, Deprecate ROI service - introduce ROI container (not on Regions branch)
- Regions v2 (Josh) Changes for first DB upgrade merged. Other work on space can continue. Josh: Where does Regions branch from BF get back into OMERO? (5.2.0?)
Bio-Formats
- Java: 5.1.5 out today(Seb in absentia)
- C++: 5.1.5 to follow (S.i.a) Superbuild tomorrow or Thursday. A couple of minor issues.
- Work on multi-version in Registry to start next week (S.i.a) Work will start in next few days
- Datasets / New formats (see below)
Glencoe Update (Chris) Upgrading to 5.1 ROI work etc.
Repositories
- Virtual Microscope/teaching
- IDR Project (see below)
FLIMfit/OPT (Ian)
- Been pulled back to FLIMfit this week - following IC sever finally being updated.
- Ready for release next week, depending on plates (new data from plate reader)
Other consortium timelines (5 minutes) ImageJ ops making good progress
IDR review/demo (20-25 minutes plus 15 mins questions max)
- Josh: for Seeing is Believing, we compiled numbers E.g. 1500 plates, 10.5 TB imported
- Sysgro: complicated / corrupt data
- Secretion screen: 900 plates - not all imported
- Getting individual plates in “painful”.
- Tara: 84 plates, 6 chambers (only 2 complete - each with 400 fields) Instead turned these into 396-well plates
- Large number of files per plate - 4 hours import
- Mitocheck: HDF5 reduces number of files but very complex data - long import times
- Melissa created screen-reader 2.0. Use multiple patterns in ini file to describe assembly of multiple files on disk into plate (extension of existing ‘pattern’ functionality).
- Do we want to support ini files in future? Is HDF file a better approach.
- 40000 readers needed - opening file can take 20 minutes.
- Propose “cacheDirectory” for Memoizer to directly access HDFCellReader
- Fixed issues (could go into 5.2.x)
- Import targets on CLI
- CLI: csv files to OMERO.tables / map annotations. Improved searching
- Various performance issues
- Other work needs planning:
- Importing into existing plate (instead of starting long import from scratch)
- Caching in web server. Too much caching (e.g. of right panel) causes bugs
- Federation of multiple servers
- Context & State - loss of selection in jsTree. Problem when working on big plates
- JM: Some other points captured during testing. Petr has notes. Context & State
- Jason: Slides of Gabriella are available.
- Different studies have never had data all in the same system
- Tara, HMS Beagle-based ocean sampling.
- 4 major studies currently in IDR - more coming.
- Ontology annotations - will be able to link out to OLS etc.
- Using current vanilla OMERO.web UI
- ‘BioStudies’ registry
- 13 Studies loaded so far. 7 on the way.
- 10.7 TB so far. 80 TB to come
- All studies have experimental metadata
- http://idr-demo.openmicroscopy.org takes you to 11 studies (screens)
- OMERO.tables metadata shown in right panel. Subset of key-value pairs picked -> Map annotations (searchable)
- Mitocheck is in different server for convenience (large) Cell h5
- 35 TB total - 10% loaded so far. More in progress
- Tara study: lots of key-value pairs on Plates - links to PDFs for experimental cards
- Vanilla web UI not optimised for various different studies but generally works
- Lots of value for these studies in querying and visualising data.
- Plan to release url to OME lists soon, then Confocal lists.
AOB (5 mins max - technical discussions should be highlighted to relevant people and rescheduled)
- Social Media review (Kelli) Currently going well, but more coordination and planning required for future E.g. announcing IDR resources Plan to start weekly meetings soon Jason: twitter engagement good. Could do more with microscope pony. Blog posts in preparation on improved BF support Become more vocal on commercial adoption?
- Conf
- Also Facility Managers’ meeting in January
-
CVs needed for grants:
- By next Tuesday.
- CVs should be in MS-Word, two pages, Arial, Font size 11
-
Holidays: keep an eye on holiday allocation - none carried over to next year
- 2 weeks off - longer needs a month’s notice
- 4pm 24th Dec - 5th Jan is University holiday period