2013 Bar Harbor GOC Meeting Agenda: Difference between revisions

From GO Wiki
Jump to navigation Jump to search
Line 28: Line 28:
#Work out what we have to do to incorporate TAS and EXP Reactome annotations in GO. Should the evidence code rules be relaxed to allow Reactome IDs in the reference field? Person: Peter D’E and Rama.
#Work out what we have to do to incorporate TAS and EXP Reactome annotations in GO. Should the evidence code rules be relaxed to allow Reactome IDs in the reference field? Person: Peter D’E and Rama.
#Form a working group to make rules for incorporating protein binding data from IntAct. Sort the QC checks and filters required. Do we want to map to Reference Genome identifiers? Person:
#Form a working group to make rules for incorporating protein binding data from IntAct. Sort the QC checks and filters required. Do we want to map to Reference Genome identifiers? Person:
== In progress from Cambridge 2013 meeting ==
# '''in progress''' (bug Seth) Goal is to figure out what should be on the public website (relevant to all kinds of users).
# '''in progress''' [Seth] move current content to the Drupal system, set up a beta site and give external users access to this. will use this time to do some spring cleaning as well.
# '''in progress''' [Seth] Set up a working group what will make the landing page.
# '''in progress''' [Seth] Seth and Paul S will work out what is useful on a stats page (via AmiGO 2).
# '''in progress''' [Seth, Chris] Seth and Chris will work with go-managers on plan for Drupal transition. Defined action items.
#Add meetings that people are attending to the central GO Calendar (Google Cal)
# '''in progress''' (but we've booted pub pages?) Once Drupal is set up then everyone (as part of the Drupal training) will update their publication
# '''in progress''' (will go with new site) [Seth] Implement CC BY as our licensing on new site
# '''in progress''' Heiko will proceed with incorporating OE user features into Protege.  He’s already added the graph viewer. No timeline for completion of new version
# '''in progress''' For terms that don’t have a template, Heiko can add a button (or some ui manipulable) to termgenie so that curators can easily submit the term request on the spot.
# '''in progress''' Work with CHEBI on providing reciprocal links from CHEBI interface (Alan Bridge’s mass spec use case)
# '''in progress'''[Chris] Chris needs to convert existing  annotation data into GPAD format that is conformant with the spec… so that protein2go will still work. Current scripts very old and need to be reviewed. Chris will make sure this is synchronized with Tony once 1.1 is defined. Make sure the ECO mappings are incorporated.
# '''in progress''' Each group (SGD, mouse, zfin, fly, worm, human) should put together some gene sets for this exercise. Mike and PaulS will shepherd this effort
== to be taken forward from Cambridge 2013 meeting ==
# '''On table''' Suzi to organize a working group on Phenotype to Process question. They will report recommendations at the April GO meeting.
# '''carried forward''' We should draft a survey and have the SAB critque the survey.
# [Rama] Rama to write the announcement, waiting for more information.
# '''Judy looking at this''', We need a automated workflow to measure what happens (can be done with Jenkins and/or Galaxy)
# '''Add to Jenkins pipeline''' Use that GOOSE query to identify and remove ISS annotations for which the supporting exp annotation don’t exist or is now contradictory. A rule for detecting transitivity errors that creep in over time..
# '''move forward''' [David] group to come up with a plan for handling chains of evidence codes and inter-ontology links.
# '''include in Paul's discussion on Scope''' Ontology editors will write down global areas of biology that needs to be developed (cleaned up). Also, we need areas that we are missing completely.
#'''include in Paul's discussion on Scope'''[PIs] set global priorities by surveying whole scope
# '''force 11 focus on publishing''' Request journal editors to require GO related data to have version # /date for when GO annotations/ontology were retrieved. What is the minimum data they need to include (PIs).


==Working Groups with Timelines==
==Working Groups with Timelines==

Revision as of 11:34, 5 October 2013

Action Items

  1. Requires_regulator; Requires_regulation_by: There are 9 annotations using these that need to be re-examined. Annotate in forward direction only. See meeting minutes for details.
  2. Set up a working group for running and displaying the term enrichment tools run on the Gold standard datasets.. Judy, Paul T. and Chris to head, Others: Mary, Paola, Ruth, Objective: They will be responsible for developing a page on the web site by the time of the March SAB meeting that summarizes an evaluation of the different enrichment tools that are available, comparing and contrasting them. Includes gene annotation gold standard sets that Rama has been shepherding. Write our own gene enrichment tool.
  3. People on GO help rota should take time to post useful things (interesting papers/topics/howto/etc.). One post per rotation will be required from everyone on the rota. Topics can be anything GO related: ontology discussion, annotation strategies, interesting papers read.
  4. PI’s to increase the number of people on the GOhelp rota.
  5. Add a user survey when we roll out the new website. The EBI folk can talk to Jenny Chan to ask for help in designing the user survey. People: ?
  6. Add IEAs for non-MODs into AmiGO2. People: Seth and Chris. In process: https://github.com/kltm/amigo/issues/48
  7. Decide what stats we want to display on the website, and which graph displays on the homepage. Send specs to Seth, current suggestions are listed in the minutes.
  8. Change settings in AmiGO2, so that Quick Search results will go to an intermediate page, displaying the various subsets (at the moment, the Annotation search page is the default, if you don’t select an option from the drop-down menu). The user can then decide which category they’re interested in. People: Seth In process: https://github.com/kltm/amigo/issues/46
  9. Add back ability to do a quick transition in AmiGO2, so you can switch quickly between annotations and terms. (requested by Midori). People: Seth In process, but needs better definition: https://github.com/kltm/amigo/issues/47
  10. FIX: Lists of gene product results are just IDs in AmiGO2. Send any text fixes or text improvements for Grebe to Seth. People: everyone.
  11. Email Heiko to request a report email on your* annotation checks. set up procedure to regularly fix. Each group to look at the Jenkins annotation report. AND FIX YOUR ERRORS please.
  12. For all MODs that represent GO, find out if representing P/F inter-ontology links would break their in-house tools. - Ask Andy Yates re EnSEMBL ayates@ebi.ac.uk
  13. Prioritise evidence codes so if looking at redundancy, propose which code(s) take priority. Required for folded annotations. People: Chris, Val, David H….
  14. Make the Jenkins-generated GAFs GPAD-compatible. Person: Heiko.
  15. Devise a mechanism for groups (like ZFIN) that like to run the checks locally before they commit their updated GAFs, etc. to the GO repository. Person: Heiko
  16. Define the checklist of all the tasks that remain to be done before the cronjob scripts at Stanford can be decommissioned and fully carried out via Jenkins. People: Heiko & Mike.
  17. Rachael to add excel file detailing the evidence code rules to the GO minutes.
  18. Check existing annotations to see if any process annotations use ‘has_direct_input’. Consider adding a rule (or a guideline, if we want to be more relaxed) to restrict has_direct_input for MF and NOT BP annotations.
  19. doneCreate a P/F link between ‘protein tyrosine kinase activity ; GO:0004713’ and ‘peptidyl-tyrosine phosphorylation ; GO:0018108’. Done. yay. (Also added link between lpeptidyl-histidine phosphorylation and protein histidine kinase activity.) Double yay.
  20. Create a subset of processes that should be excluded from enrichment analyses/slims. Person: Paul T.
  21. Create a TG template for protein complexes.
  22. Change relationship in physical interaction terms (logical definitions) to be has_direct_input instead of has_input. See minutes for details.
  23. Improve documentation on how ‘regulates’ relationships should be used for BP and MF: when you should (and should not) propagate over ‘regulates’ relationships. People: Chris.
  24. Either standardize AmiGO2 and QuickGO so the default for propagating over ‘regulates’ annotations in BP is the same. Or (less preferred) is to document the tool behaviour and display VERY CLEARLY what the default search will retrieve. People: Chris, Seth.
  25. Form a working group to come up with how we’re going to handle predicted annotations submitted by external groups. We need to come up with QC and standards, quality metrics, and which evidence code is appropriate: all PIs, Susan, Donghui, Rachael, Val.
  26. Work out what we have to do to incorporate TAS and EXP Reactome annotations in GO. Should the evidence code rules be relaxed to allow Reactome IDs in the reference field? Person: Peter D’E and Rama.
  27. Form a working group to make rules for incorporating protein binding data from IntAct. Sort the QC checks and filters required. Do we want to map to Reference Genome identifiers? Person:


In progress from Cambridge 2013 meeting

  1. in progress (bug Seth) Goal is to figure out what should be on the public website (relevant to all kinds of users).
  2. in progress [Seth] move current content to the Drupal system, set up a beta site and give external users access to this. will use this time to do some spring cleaning as well.
  3. in progress [Seth] Set up a working group what will make the landing page.
  4. in progress [Seth] Seth and Paul S will work out what is useful on a stats page (via AmiGO 2).
  5. in progress [Seth, Chris] Seth and Chris will work with go-managers on plan for Drupal transition. Defined action items.
  6. Add meetings that people are attending to the central GO Calendar (Google Cal)
  7. in progress (but we've booted pub pages?) Once Drupal is set up then everyone (as part of the Drupal training) will update their publication
  8. in progress (will go with new site) [Seth] Implement CC BY as our licensing on new site
  9. in progress Heiko will proceed with incorporating OE user features into Protege. He’s already added the graph viewer. No timeline for completion of new version
  10. in progress For terms that don’t have a template, Heiko can add a button (or some ui manipulable) to termgenie so that curators can easily submit the term request on the spot.
  11. in progress Work with CHEBI on providing reciprocal links from CHEBI interface (Alan Bridge’s mass spec use case)
  12. in progress[Chris] Chris needs to convert existing annotation data into GPAD format that is conformant with the spec… so that protein2go will still work. Current scripts very old and need to be reviewed. Chris will make sure this is synchronized with Tony once 1.1 is defined. Make sure the ECO mappings are incorporated.
  13. in progress Each group (SGD, mouse, zfin, fly, worm, human) should put together some gene sets for this exercise. Mike and PaulS will shepherd this effort


to be taken forward from Cambridge 2013 meeting

  1. On table Suzi to organize a working group on Phenotype to Process question. They will report recommendations at the April GO meeting.
  2. carried forward We should draft a survey and have the SAB critque the survey.
  3. [Rama] Rama to write the announcement, waiting for more information.
  4. Judy looking at this, We need a automated workflow to measure what happens (can be done with Jenkins and/or Galaxy)
  5. Add to Jenkins pipeline Use that GOOSE query to identify and remove ISS annotations for which the supporting exp annotation don’t exist or is now contradictory. A rule for detecting transitivity errors that creep in over time..
  6. move forward [David] group to come up with a plan for handling chains of evidence codes and inter-ontology links.
  7. include in Paul's discussion on Scope Ontology editors will write down global areas of biology that needs to be developed (cleaned up). Also, we need areas that we are missing completely.
  8. include in Paul's discussion on Scope[PIs] set global priorities by surveying whole scope
  9. force 11 focus on publishing Request journal editors to require GO related data to have version # /date for when GO annotations/ontology were retrieved. What is the minimum data they need to include (PIs).

Working Groups with Timelines

  1. Working group for running and displaying the term enrichment tools run on the Gold standard datasets. Judy, Paul T. and Chris to head, Others: Mary, Paola, Ruth, Objective: They will be responsible for developing a page on the web site by the time of the March SAB meeting that summarizes an evaluation of the different enrichment tools that are available, comparing and contrasting them. Includes gene annotation gold standard sets that Rama has been shepherding. Write our own gene enrichment tool.
  2. Working group to come up with how we’re going to handle predicted annotations submitted by external groups. We need to come up with QC and standards, quality metrics, and which evidence code is appropriate: all PIs, Susan, Donghui, Rachael, Val.
  3. Working group to make rules for incorporating protein binding data from IntAct. Sort the QC checks and filters required. Do we want to map to Reference Genome identifiers? Person: include Sandra Orchard

Pre-Meeting Working Groups

Thurs 3 Oct

These pre-meeting working groups will meet at The Jackson Laboratory. Please arrive 30 minutes before the session you wish to attend so you can get a badge and escort.

  • 9:30am - 10:30am Web documentation, Drupal transition (room ERB 1421)
  • 11:00am - 12:00pm Ontology group discussion: OWL/OBO editing workflow (room NRB 3035)
  • 12:00pm - 1:00pm Lunch at Roscoe's
  • 12:50pm - 1:30pm Lab tour (optional; Mark Wanner leading)
  • 1:30pm - 2:30pm Annotation extensions: (room ERB 2519)
    • for those already making c16 annotations and ontology developers; This group will be finalizing what they want to present to the whole GOC during this meeting
  • Annotation tools (the following discussions are all interrelated, ordering here is the same as in main agenda)
    • 2:30pm - 3:00pm LEGO group discussion (room ERB 2519)
    • 3:00pm - 4:00pm PAINT group discussion (room NRB 3035)
    • 4:00pm - 5:00pm CAT working group: next steps moving from Protein2GO to free-standing CAT, independent of UniProt (room NRB 3035)

Agenda

Minutes here: https://docs.google.com/document/d/1JuQXt_7zgb_3wkFnsz4Lqqm0gy0UvbH3hBM4unse8FI/edit?pli=1

The SAB report from April 2013 (Cambridge) was very specific and productive [1]. We will be meeting with them again in March 2014 and need to show progress on their useful recommendations. Therefore the focus of this meeting is to identify targeted projects that we can demonstrate to them by completely them in the (roughly) six intervening months. Establishing priorities, point people who will handle coordination, and monthly milestones.

Friday 4 Oct - Annotation Impact & Usability

Breakfast on your own

8:30 am start (lunch 12:30-2:00 pm, breaks at 10:30 am and 3:30 pm All EDT)

  1. Assign 'official' note takers, set up Google doc for notes and so everyone can augment, introductions of new people
  2. 9:00 - 9:30 am
    • GO-pal transition (GO + Drupal = GO-pal), Seth + Moni
    • GO annotation stats Seth
    • Term enrichment front & center
      • Gold Standard Annotation Sets Rama
      • Biological data interpretation (aka enrichment) Chris
    • Update on tool registry Chris
    • GO analysis gotcha page (e.g. post summary of PLoS article on close/open world?) Suzi
  3. 9:30 - 10:00 am AmiGO 2 Seth
    • Demo with David H David
    • Display of external ontologies in AmiGO 2 Chris
    • Timeline for transition to Stanford hosting
    • Filtering feature requests
      • Could we see the major models in the first set, rather than broad taxon groupings?
      • Maybe the "text box" filter should be available without going to "more..."
  4. 10:00 - 10:30 am Engaging the user community (GO help responsibilities) Suzi
    • A Gene Ontology MOOC (Coursera, EdX, Udacity, etc.) to train and to answer "I want to ..." questions
    • Ideas for dynamic content: e.g. GO help rota responsibilities would include writing blog on latest annotation paper that was discussed, or one of the help questions.
    • GO-FAQ and GO-help Chris

BREAK: 10:30 - 11:00

  1. 11:00 - 11:30am Annotation workflow (Quality control reports and how to manage them efficiently)
    • Review of Jenkins GAF QC checks and inferences
      • Location of files and what they contain - Heiko
      • Examples from the MGI file and discussion of most efficient ways to deal with reports David H
    • With/From column IDs (what is mandatory, how to move this to Jenkins for checking) Rama/Rachael
  2. 11:30 - 12:30 pm Tutorial on column 16 Rachael
    • Proposed json file format for GO annotations (up to LEGO-expressivity) Chris

LUNCH: 12:30 - 2:00 pm

2:00 - 3:00 pm

  1. Column 16 examples: tutorial extension Rachael
  2. Annotation procedures Val
    • Differences in how we treat annotation propagation the regulation relationship for function and process (implications for Ensembl) implications of not propagating over "regulates" e.g;metaphase anaphase transition (0) regulation of metaphase/anaphase transition (51), if regulation does not propagate (as in Ensembl) other e.gs regulation cell cycle/ regulation of mitotic cell cycle

BREAK: 3:30 -3:45 pm

3:45 - 4:45pm

  1. Incorporation of external annotations.Judy
    • Discuss which evidence codes to use for annotation predictions.
    • Discuss strategies for incorporation, including non-MOD IEAs
    • Reactome annotations, their use of ev.codes etc File:ProteinBindingGOC2013.pdf
    • Proposal to integrate protein binding annotations


Conference Dinner at 5:30 pm Cafe This Way

Saturday 5 Oct - Ontology & Infrastructure

Breakfast on your own

8:30 am start (lunch 12:30-2:00pm , breaks at 10:30am and 3:30pm EDT)

8:30 - 9:00 am

  1. Ontology development infrastructure update, transition to OWL, etc. Chris

9:00 - 10:30 am

  1. Ontology developments Judy
    • Updates on content in specific biological domains
      • Neuronal cell components and integration with NIF Paola
      • Apoptosis annotation update Paola
      • Proposed Giardia/unicellular eukaryote cc expansion Paola
      • cell cycle Jane
      • virus project update Becky
      • Uberon/CL/GO developments David OS

BREAK 10:30-11:00 am

11:00 am -12:30 pm

  1. Molecular Function alignment with Rhea Chris
  2. Biological Modeling (LEGO) Paul T
    • Modular annotation ("lego") progress report Chris

LUNCH 12:30 - 1:30pm

1:30 - 3:30 pm

    • Reactome Discussion
      • Automated translation of Reactome & BioPAX to LEGO/OWL Chris
      • Glycolysis example for comparison David H
      • Pathways in GOC, representation and prioritization
      • Relationship between GOC and various pathway representations including wikipathways, using these external definitions to define the starts and ends of slim processes

3:30 pm Group Picture

3:30 - 4:00 pm BREAK

4:00 - 5:00 pm

  1. Annotation Infrastructure Suzi
    • PAINT annotation and tool update Paul T & Suzi
    • Common Annotation Tool update
      • SGD's Migration to Protein2GO Rama
      • Integration of Text Mining Tools with Protein2GO Kimberly
      • CAT: beyond Protein2GO PaulS


Field Trips: Enjoy Acadia National Park. 5:00pm

Sunday 6 Oct - Summary and planning

  1. Parked items
    • More guidelines about when to precompose/not precompose (especially using during/involved in response to)
  2. Action Item review and assignments
  3. Monthly milestones for next 6 months
  4. Future Meetings
    • LEGO Hackathon
    • Glycoproteins - content and annotation workshop
    • GOC meeting next Spring (Jim Hu, Paul Thomas) (College Station, Texas) will include SAB meeting
    • Possibility of an Annotation Camp in the near future
  5. Papers published and in progress