Manager 03September2014

From GO Wiki
Jump to navigation Jump to search

On call:

Trello Board Link:

Additional Agenda Items

  • ECO development? It has been very slow.


ECO development

  • Rama: protein2go displays ECO code for evidence. Currently Marcus is the only person adding new terms and the process is slow.
  • David: Is the project still funded?
  • Pascale: Submitted 20+ terms with definition back in June. Still haven’t heard anything back from Marcus.
  • Suzi: Continue to work with Marcus.
  • Chris: We should all be able to commit terms to ECO. Chris will give write permission to Pascale, anybody else. Marcus can review the terms later.
  • Rama: We also need a mechanism to alert curators about new ECO codes so we know what was recently added (similar to SF digests or Jenkins reports). Chris thinks this can be done via Google code.

Trello board updates

  • Suzi: Trello board - anything needs to be discussed, not completed in the same day, should be added to the Trello board. Take it off when it is completed.

Ontology update (David)

  • Triage SF items - Paola almost finish the sourceforge minutes.
  • Protege testing - It is working. Still use OBO-edit with OBO file. For more complicated file, convert it to owl file and open it with Protege, and then convert back to OBO.
    • Chris asked why not using Protege all the time.
    • David: OBO editor has better visualization, and is easier to use.
  • pattern/template development - get reasoning to work in part-of inference. Sometimes it is difficult, eg., myocardium is part_of heart, but development of myocardium is not part of development of heart.
  • Assess PT's plan on MF change - David OS is working on it.
  • Cell cycle ontology fixes - Jane introdices phases of cell cycle.. pretty close to be completed.
  • protege5 - switch soon
  • protein binding - continue to work on it.
  • small protein - rama will update at the next call. ubiquitin ligase
  • LEGO example - like to use noctura for col 16 data. problem is where to get resource for that.
  • Judy - what is the highest priority?
  • David: continue to get sourceforget items down, and get
  • Suzi - the order of the card should reflect priority. come off when it is done.

Common annotation framework (Chris)

  • Notura - beta test released. Can create a title to a new model. Create a new model. Working on LEGO example.
  • Authentification - is working now. Take it off from the board.
  • Rama-enrichment tool lacking some of the nice feature. When will be improved.
  • Chris - soon.

Outreach and usability, Annotation Advocacy(Rama)

  • GO help - average 1 email a day. We get really good questions from our users. There was a bug in our GO help email receiving process (2 emails did not make it to the JIRA). Hope we can prevent it from happening in the future.
  • google analytics - Amigo - 20K sessions first month, 8000 users
    • Suzi: Anyway to tell internal from external?
    • Rama: Not yet.
  • Will show full results at the barcelona meeting
  • Moni has made a video for the GO website.
  • Annotation advocacy - documents for new groups to create annotation, but has not trained new annotation groups in the last quarter.
    • Rachel - Prudence will give training to a group in Brazil interested in annotating blood/brain barrier genes?
    • We are working improving the documentation for IPI
    • Groups are making progress towards cleaning up un_published references
  • RCA annotation - review and remove.
    • Chris- has a script that can compare a RCA annotation with a manual and allow curators to decide if it is worth keeping the RCA.
    • Judy - still need manual review to see if there are new evidences.
    • Paul T. - a script may help find new evidences.

PAINT curation (Huaiyu)

  • Jamboree still going, but slow. Marc is the biggest contributor.
  • New PANTHER build to support the current version of reference proteome is underway.
    • The targeted release date is the end of year.
    • It will have around 100 genomes, which includes the current 86 genomes, and 20 new plant genomes. However, we will remove a few low quality genomes. Suzi may be able to provide a few examples.
    • Suzi: Only see it in a limited number of families. What if it is bad in the family, but fine in the others?
    • Huaiyu: We will compile the list, and probably use a script to check across the entire library.
    • Suzi: We should communicate this to the reference proteome team.
    • Paul T.: Yes this is the plan.