2018-09-18

Dundee: N/A

Remote: Mark, Josh, Seb, June, Simon, Wilma, Melissa, Kenny, J-m, Frances,
Chris, Will, David, Riad, Emil

Start: 2:00 pm

1. Accepting minutes from last meeting

2. Project Status

(2-3 minutes each)

• 5.4.8 release (Josh)

  • Bug found today - saving rdef in 5.4.8 Insight read-only. View

    thumb in web.

  • Chris: read-only group often used for publishing. E.g. IDR.

    • Some existing issues in 5.4.7.

    • New edge-case bug may not be a blocker.

    • Need to review in more depth after.

    • HMS - plate thumbnails stopped working. Thumb versions out

      of sync.

    • Fixed by resetting. Source of problem not known.

  • Riad: seems I have a fix for current bug.

  • Josh: OK, lets retest etc.

• Epics

  • Training (JM):

    • Neubias: Focus on Matlab. PRs open. Testing volunteers?

      • Don’t know which version will be used at Neubias

        • Currently testing the scripts with 2017a

      • Our ci uses older version - doesn’t support latest

        scripts

    • IDR training changed

      • JM will continue VAE notebooks cleanup

• Glencoe (Chris):

  • Working on parade workflows - Increased focus on parade next

    week.

    • JM new version available end of October for Japan? Chris:

      probably.

  • FileFormats - czi diverse formats

  • Pyramids work - PR opened.

• Community

  • HCA (Josh):

    • File formats: LIF is joining ND2

    • Visualization uses cases (see #hca channel for summary)

      • Raw images from scopes - convert to starfish format.

        Modulo.

      • Asked for analyzed data (spots, ROIs etc).

      • 1000’s of genes per image. E.g. which planes have gene

        ROI on.

      • Do we use ROI folders? Masks, annotations etc?

      • Mentioned figure - easy win to encourage usage.

• Sysadmin roundup:

  • Data_repo filled-up squig.

    • Check_MK python script to alert this in future
  • GPFS network memory settings causing problems. Reverted changes

    has resolved this.

  • Kubernetes production failure. Master and Nodes now in Check_MK

    for CPU, RAM, etc, and Check_MK (local check) script now on Kubernetes master to check “Ready” status of all 3 nodes.

  • External security testing from UoD IT/JISC not found any

    critical issues in OME

1. AOB

(5 min. max; tech. Discussion should be highlighted to relevant people and rescheduled)

1. Main Topic

(20-25 minutes plus 15 minutes questions max)

• Upcoming efforts: Data-In & Data-Out (Seb)

  • Data-In:

    • Responsibilities: IDR datasets + ROIs, HCA datasets + ROIs

    • Technologies: omero-roi(-py,-java),

      biotracks, Bio-Formats 6, S3Reader

    • Team: David, Dom, Frances, Josh, Kenny, Seb, Simon

  • Data-Out:

    • Responsibilities: training -> ITR, 3rd party integration

      (e.g. Knime) Matlab, visualization

    • Technologies: apps, microservices, downloader

    • Team: J-m, Mark, Petr, Riad, Will

• Notes

  • Seb: Data-in includes images, metadata, rois, etc

  • Slides upcoming on each of the fronts for general clarity

  • Less a focus on software (when not driven by another

    deliverable)

  • Josh: matches what was discussed during OSARs:

    • IDR is driving force for new features, UI viz etc.

  • J-m: also matches what we’ve been doing in the training sessions

    • Note issues of large exports

  • Reporting back over the next week? (despite getting 5.4.8 out)

    • Also need to define interactions with Glencoe (parade,

      pyramid)