Personal tools
  • We're Hiring!

You are here: Home Community Minutes Conference Calls 2015 2015-12-08 Tuesday Team Meeting

2015-12-08 Tuesday Team Meeting

Dundee: Will, Kenny, Gus, Petr, Dominik, June, Helen, Simon, Jean-Marie, Colin, Ola, Roger, Simone

Remote: Ian, Josh, Sebastien, Balaji, Andreas, Melissa, Kelli, Emil,

Agenda - 2:30pm Start

  1. Accepting minutes from last meeting

  2. Project Timelines (2-3 minutes each)

    1. OMERO

      1. 5.2.1 (J-m) PR freeze yesterday. Start merging tomorrow after review Centos6 Python2.7 docs testing ongoing Won’t have 5.2 docs for Centos6 & python 2.6 - can refer to 5.1 docs if needed Pillow 3.0.0
      2. Metadata (Josh) Hinxton: “what is metadata” EBI are working on submission tool to handle all datasets from everyone Jason suggested hackathon to hack on this. Image reader / writer to handle other data types.
    2. Bio-Formats

      1. 5.1.7 (Melissa)
        • hopefully out by now - (yesterday)
      2. focus on 5.2.0 now, no plans for 5.1.8
      3. 5.2.0 (Josh/Seb)
        • Mark: initial BF/OMERO work pushed to "folders" branches
        • Josh: passing ownership of regions branches to Seb to migrate to mainline.
      4. Seb: almost rebased all 5.1.x now (finish today)
      5. Then integrate regions branch - develop branch will be unstable
      6. Shapes (David), ROI folders (Mark), integration with OMERO all next year
      7. Data repository: “from skyking” is now “curated”
      8. Petr: will keep “test_images_good”.
  3. Create “test_images_mtools” which will link to “curated” folder

  4. “inbox” folder to drop new images for discussion in bio-formats standup

    1. Other releases/upgrades:

      1. Figure PR with small changes to build process JM: planning workshop on figure in mid January for those interested
      2. FLIMfit/OPT (Ian) On backburner just now Mass spec work. What do we show to whom & when? Need Jason’s input.
      3. IDR Project (Eleanor/Josh) DOIs ( ), 11th deadline (“phase 2”) Simon: we have new servers & access to EBI’s cloud (openstack) Need new naming convention for servers! Josh: 3 phases - Freeze data on friday - lock down access etc. DOI above from Dundee uni points to idr demo. Then distribute data to EBI’s cloud in January. Funding renewal… Eleanor: trying to get several more datasets in by Friday. A bunch more for January. JM: need to analyse idr demo logs to determine public use of repo
      4. ImageJ JM: No changes for 5.2.1 Curtis: Alison is getting up to speed with OMERO since last Tuesday. Her first priority will be merging the old tables branch to imagej-omero master. We also updated to OMERO 5.2. Curtis: Java 8: we are continuing to push forward. The Java 3D woes are mostly (but not fully) solved. Java compiler woes are being solved now by Mark. We also updated ImageJ Ops master branch to use Java 8 now, and it wreaks havoc with ImageJ startup. I am patching various problems to ease the migration there.
      5. Learning (Ola) Ray updated to 5.2.0 Petr tested - a couple of small web issues. Move to haddock (ray is still on ice 3.4)
      6. Sysadmin Upgraded Demo to latest figure and Pillow 2.9.0 SLS provided servers for IDR (see above) Roger: server for VMware? Kenny: potentially. Josh: Openstack instead of VMware?
    2. Glencoe Update (Chris) Pathviewer launched - went well. Digital Pathology meeting in London. Pathviewer received well Marshalling work ongoing

  5. Presentation on localisation by Ian Munro

    • Localisation microscopy intro:
    • Resolution limited by size of Airy disk - wavelength dependent
    • Add fluorescent proteins (very small point source) - Can pinpoint centre of each airy disk.
    • With low density of fluorescent points can identify point sources.
    • Various ways of ensuring only a small proportion of fluorophores are active at a time
    • Turning them on/off in turn.
    • PALM, STORM etc.
    • Analysis of stack of raw images, to generate high resolution output image
    • No standard of file formats or metadata.
    • Various different reconstruction algorithms for localisation of fluorophores and image reconstruction.
    • From OMERO point of view - image data can be handled easily.
    • Lacking format for storing point localisations.
    • E.g. Stack of 15000 timepoints 3k * 3k image -> single analysed image
    • Simon wrote script for taking localisation data -> OMERO.table
    • Demo: PALMsiever matlab tool - pick Dataset from OMERO - with table attached.
    • Generates high resolution image
    • ThunderSTORM ImageJ plugin. Ian has added import options from OMERO.
    • Pick Dataset as before, also generates high res image
    • Both approaches use UI widget from FLIMfit for picking Projects and Datasets from OMERO.
    • Continue with this approach or use other options available
    • Investigate other options for storing table data.
    • Do both tools need a way of converting data from different formats into OMERO.table?
    • Josh: Reader could create tables on import.
    • JM: Specific namespace for table data, could allow ImageJ OMERO plugin to open this in thunderSTORM (instead of logging in from thunderSTORM)
    • Balaji: Need memory management to avoid problems during analysis.
    • Ian: Curtis suggested using omero imageJ (server-side analysis based on image ID).
    • JM: configuration of hard-coded demo parameter for loading subset of data?
    • Ian: lots of features are still missing - still at proof of concept stage
    • JM: how many file formats are there
    • Ian: about 40?! Tools above can handle about 5 each, but no format is supported by both.
    • Ian: PALMsiever changes have been merged into main-line (unsure if reviewed much)
    • ThunderSTORM devs not been made aware of this work yet.
    • JM: Do we have samples for the different formats?
    • Ian: No - some format examples may be collected by someone? need to check
    • JM: can you make the slides available. Ian: OK, will add screenshots too.
  6. AOB (5 mins max - technical discussions should be highlighted to relevant people and rescheduled) Microscopy managers’ meeting in Oxford 7th & 8th Jan (Jason & Simon to attend)

Document Actions