GO Advisors Meeting: Difference between revisions

From GO Wiki
Jump to navigation Jump to search
mNo edit summary
 
(25 intermediate revisions by 6 users not shown)
Line 1: Line 1:
==Meeting Room==
The GO Advisors Meeting  will be held in Conference Room 316L of the South Campus Center of the University of Washington.  There will be transportation from the Silver Cloud Inn to the Conference Center starting at 7 am.  The meeting will start at 8:00 am.
== Aim of the meeting  ==  
== Aim of the meeting  ==  
The main focus of this meeting is to obtain our advisors views and perspectives on the work we have planned (i.e. the new proposal) and the approach we are taking to accomplish our aims.  
The main focus of this meeting is to obtain our advisors views and perspectives on the work we have planned (i.e. the new proposal) and the approach we are taking to accomplish our aims.  
'''''Topics'''''


Please add items below that you think need to be presented. When possible, please put your request in priority rank order:
Please add items below that you think need to be presented. When possible, please put your request in priority rank order:


'''''Utility for Biologists'''
'''''Utility for Biologists'''''
 
The highest priority for the GOC identified by grant reviewers and others is to enhance the ability of non-informatics biologists to access the GO resources as part of their data analysis efforts.  Some of the participants in the GO Consortium provide access to GO data analysis tools as part of their MOD resource.  But the GO site itself does not support data analysis directly from the site. We do provide listing of tools, but have not seen it as part of our 'mission' to support a general tool [such as GO TermFinder, for example] as part of the GOC. Should we do so?  Wouldn't this be analogous to provide a BLAST server?  While the individual MODs have a role to play here, the GOC is seen by some to be behind in supporting the work of the bench biologist.
The highest priority for the GOC identified by grant reviewers and others is to enhance the ability of non-informatics biologists to access the GO resources as part of their data analysis efforts.  Some of the participants in the GO Consortium provide access to GO data analysis tools as part of their MOD resource.  But the GO site itself does not support data analysis directly from the site. We do provide listing of tools, but have not seen it as part of our 'mission' to support a general tool [such as GO TermFinder, for example] as part of the GOC. Should we do so?  Wouldn't this be analogous to provide a BLAST server?  While the individual MODs have a role to play here, the GOC is seen by some to be behind in supporting the work of the bench biologist.
* I think this should be given a high priority for discussion. In the AmiGO WG we've discussed adding simple tools (e.g. a cgi of map2slim) to AmiGO, but for some of the more complex tools e.g. microarray tools it's probably not worth developing our own as so many good tools already exist. But should we 'endorse' some of these existing tools? And then how could we have say in the direction they're developed? We already work fairly closely with some of these groups (Onto-Express, GoMiner).
* Related to the list of tools, its great to have this long list but it makes it very hard for new users to know which to use and why. Some sort of critical review of features and suggested uses would be very handy. Selecting/Endorsing a 'best' tool in a given category (if there was/were clear winners) would also be really helpful to users but maybe more controversial if done by GOC vs some external author.
** In reviewing the tools one would identify lists of essential and desirable features that 'good' GO tools should ideally possess. If the GOC could approve a set of core features that all 'endorsed' tools should have this might provide a way to influence future tool development by others. Tools that met or exceeded the feature set would get higher billing than ones which didnt, presumably driving non-compliant tools to meet these higher standards or go by the wayside (similar to OBO foundry driving the development of higher quality ontologies).
*In general, how much do we know about how bench biologists use GO? What tools and other enhancements would make biologists more aware of what GO has to offer, and help them use our resources? What do they want to do with GO?
'''''Utility for GO-Tool Developers'''''
This is a note from a developer...
"we are in the process of making improvements to our GO analysis tool, the
Ontologizer, and I have noticed that one of the most time consuming
aspects of the process seems to be finding good datasets with which to
test new algorithms. I would guess that we are not entirely alone in this.
I wonder if it would be a good idea for the GO community to make a
repository of datasets for testing new algorithms and also for new users
to learn the ropes. I would be thinking of lists of study sets (say
differentially expressed genes) with corresponding population sets (say
all genes on the chip) together with short biological and methods
descriptions. Does such a repository already exist or does anyone have any
tips?"
'''''Should GO capture the context of activity'''''
Is there a facility to construct GO annotations which would capture the context in which a given gene product is associated with (e.g.) a given function? Thus, for example, instead of annotating protein X to oxygen transport [function] one might annotate:
<protein X in cell component Y> to oxygen transport [function']
For MGI, this information is captured for cell types, anatomy and target molecules, but it is not a GO-wide policy. MGI captures this information in structured notes. So, for example MGI curators can make an annotation for gene product X having some function in oxygen transport in a lung epithelial cell. The hope is that if X is also annotated to oxygen transporter activity in lung epithelial cell and plasma membrane of a lung epithelial cell, then we can draw a conclusion that X has the potential to execute a function of an oxygen transporter involved in oxygen transport in the plasma membrane of a lung epithelial cell. In the right spatio-temporal context, the function will be executed.
'''''Encouraging community annotation'''''
#How can we help community scientists to get credit for their GO work? Could we publish a short account of the work of individual content meetings with all experts having author credit? Could groups providing manual annotation of their own gene products of interest get a small publication giving an account of their contribution and of possible applications?


'''''Annotation Tool'''''
'''''Annotation Tool'''''
Line 11: Line 50:
#How does the idea of an annual functional annotation tool bake-off sound to you? How would you approach it? Can it be used to keep orphaned genomes up-to-date?
#How does the idea of an annual functional annotation tool bake-off sound to you? How would you approach it? Can it be used to keep orphaned genomes up-to-date?
#Almost everyone agrees that the GO Consortium should develop an annotation tool, [but not all people since most groups do annotation within the context of their genome annotation stream] Once we begin asking questions about what the tool should do, it becomes clear that every group has very different ideas about what such a tool should do. Groups within the GO tend to advocate for the development of a tool that satisfies their own research efforts at the moment, because that is the immediate need as they understand it. Could the advisors suggest a way of determining which of these special-purpose applications are best for the GO user community ''as a whole'', so that we can decide how to focus our development efforts? What's the best way to encourage inter-group communication to foster collaboration on tools and minimize redundant development efforts? How can we decide whether specific, highly targeted annotation tools or general, lowest-common-denominator annotation tools would be most useful to our user community?
#Almost everyone agrees that the GO Consortium should develop an annotation tool, [but not all people since most groups do annotation within the context of their genome annotation stream] Once we begin asking questions about what the tool should do, it becomes clear that every group has very different ideas about what such a tool should do. Groups within the GO tend to advocate for the development of a tool that satisfies their own research efforts at the moment, because that is the immediate need as they understand it. Could the advisors suggest a way of determining which of these special-purpose applications are best for the GO user community ''as a whole'', so that we can decide how to focus our development efforts? What's the best way to encourage inter-group communication to foster collaboration on tools and minimize redundant development efforts? How can we decide whether specific, highly targeted annotation tools or general, lowest-common-denominator annotation tools would be most useful to our user community?
#This thread is
'''''Bold text''Encouraging community annotation'''
#How can we help community scientists to get credit for their GO work? Could we publish a short account of the work of individual content meetings with all experts having author credit? Could groups providing manual annotation of their own gene products of interest get a small publication giving an account of their contribution and of possible applications?


== Agenda ==
== Agenda ==
In order to meet our objectives, we need to present all of the material in the morning.
In order to meet our objectives, we need to present all of the material in the morning.
The presentations will focus on the goals and progress of the Consortium and will reflect the new organization and management of the GOC developed through the process of writing the GO Consortium competitive renewal.
The presentations will focus on the goals and progress of the Consortium and will reflect the new organization and management of the GOC developed through the process of writing the GO Consortium competitive renewal. The times noted here are expected to be divided between a short presentation and then focused discussion on the topic.
 


Morning presentations will be 30 minutes, plus 15 minutes for questions
** - 8:00AM: Introductions (picture on display and circle everyone)
** - 8:10 AM: Overview of Grant Aims (Michael-20 minutes)
** - 8:30 AM: Reference Genomes (Rex - 30 minutes)
** - 9:00 AM: Ontology/Database Developments - Biological (Midori and David-30 minutes)
** - 9:30 AM: Ontology Developments – Technical (Chris-30 minutes)
** -10:00 AM: Break (15 min)
** -10:15 AM: Emerging Genomes (Jen-20 minutes)
** -10:35AM:  User Advocacy including AmiGO (Eurie & Jane- 20 minutes)
** -10:55AM:  New management structure (Suzi- 20 minutes)
** -11:15AM:  Summary of issues we face: pink sheets and more (Judy- 20 minutes)
** -11:35AM: Open Discussion (all)
** -12:30PM – 1:30PM, working lunch for Advisors and GO-top


The workshop will consist of these sessions:
** - 1:30PM - 3:00PM, Advisors powwow
*GO presentations: New initiatives in the renewal
** - 3:00 PM - break
**8:00AM: Introductions (picture on display and circle everyone)
** - 3:30PM - 5:00PM, feedback from Advisors
**8:15AM: Ontology Developments (technical - Chris; biology - David & Midori)
** - 6:00PM dinner with everyone
**9:00AM: Reference Genomes (Michael & Rex)
**9:45AM: Outreach (Jen)
**10:30AM: 15 minute Break
**10:45AM: Advocacy (Eurie & Jane)
**11:30AM: New management structure (Suzi)
**12:15PM: Summary of issues we face: pink sheets and more (Judy)
*1:00PM - 2:00PM, working lunch for advisors and GO-top
*2:00PM - 3:30PM, advisors powwow
*3:30PM - 5:00PM, feedback from advisors
*6:00PM dinner with everyone


== Participants ==
== Participants ==
Line 52: Line 89:
#Simon Twigger
#Simon Twigger
#Tanya Berardini
#Tanya Berardini
#Nicky Mulder
*GO Advisors
*GO Advisors
#<b>Larry Hunter</b>
#<b>Larry Hunter</b>
Line 61: Line 99:
#<b>Peter Tarczy-Harnach</b>
#<b>Peter Tarczy-Harnach</b>
#<b>Ian Dix</b> (Courtland Yockey attending)
#<b>Ian Dix</b> (Courtland Yockey attending)
*NIH Representative
#Peter Good
*Other possible attendees
*Other possible attendees
#Monte Westerfield?
#Monte Westerfield?
#Peter Good?


== Venue ==
== Venue ==
Line 71: Line 110:
   Seattle, WA 98105
   Seattle, WA 98105
   Phone: 206-526-5200
   Phone: 206-526-5200
== Slides ==
* [[Media:20060401_ProjOrg_Suzi.ppt]]
* [[Media:20060909_SeattleSAB.ppt]]
* [[Media:AnnotationOutreachnosheep.ppt]]
* [[Media:GO Advisory Committee Reference Genomes.ppt]]
* [[Media:GO User Advocacy_v5.ppt]]
* [[Media:GO-User-Advocacy-final.ppt]]
* [[Media:GOadvbd_dph06.ppt]]- [[Media:QuestionsForSAB.doc]]
* [[Media:adv_board_2006-MAH.ppt]]
* [[Media:go-sw-for-advisors-2006.ppt]]
[[Category:GO SAB Meetings‏‎]]

Latest revision as of 12:21, 13 April 2019

Meeting Room

The GO Advisors Meeting will be held in Conference Room 316L of the South Campus Center of the University of Washington. There will be transportation from the Silver Cloud Inn to the Conference Center starting at 7 am. The meeting will start at 8:00 am.

Aim of the meeting

The main focus of this meeting is to obtain our advisors views and perspectives on the work we have planned (i.e. the new proposal) and the approach we are taking to accomplish our aims.

Topics

Please add items below that you think need to be presented. When possible, please put your request in priority rank order:

Utility for Biologists

The highest priority for the GOC identified by grant reviewers and others is to enhance the ability of non-informatics biologists to access the GO resources as part of their data analysis efforts. Some of the participants in the GO Consortium provide access to GO data analysis tools as part of their MOD resource. But the GO site itself does not support data analysis directly from the site. We do provide listing of tools, but have not seen it as part of our 'mission' to support a general tool [such as GO TermFinder, for example] as part of the GOC. Should we do so? Wouldn't this be analogous to provide a BLAST server? While the individual MODs have a role to play here, the GOC is seen by some to be behind in supporting the work of the bench biologist.

  • I think this should be given a high priority for discussion. In the AmiGO WG we've discussed adding simple tools (e.g. a cgi of map2slim) to AmiGO, but for some of the more complex tools e.g. microarray tools it's probably not worth developing our own as so many good tools already exist. But should we 'endorse' some of these existing tools? And then how could we have say in the direction they're developed? We already work fairly closely with some of these groups (Onto-Express, GoMiner).
  • Related to the list of tools, its great to have this long list but it makes it very hard for new users to know which to use and why. Some sort of critical review of features and suggested uses would be very handy. Selecting/Endorsing a 'best' tool in a given category (if there was/were clear winners) would also be really helpful to users but maybe more controversial if done by GOC vs some external author.
    • In reviewing the tools one would identify lists of essential and desirable features that 'good' GO tools should ideally possess. If the GOC could approve a set of core features that all 'endorsed' tools should have this might provide a way to influence future tool development by others. Tools that met or exceeded the feature set would get higher billing than ones which didnt, presumably driving non-compliant tools to meet these higher standards or go by the wayside (similar to OBO foundry driving the development of higher quality ontologies).
  • In general, how much do we know about how bench biologists use GO? What tools and other enhancements would make biologists more aware of what GO has to offer, and help them use our resources? What do they want to do with GO?

Utility for GO-Tool Developers

This is a note from a developer... "we are in the process of making improvements to our GO analysis tool, the Ontologizer, and I have noticed that one of the most time consuming aspects of the process seems to be finding good datasets with which to test new algorithms. I would guess that we are not entirely alone in this. I wonder if it would be a good idea for the GO community to make a repository of datasets for testing new algorithms and also for new users to learn the ropes. I would be thinking of lists of study sets (say differentially expressed genes) with corresponding population sets (say all genes on the chip) together with short biological and methods descriptions. Does such a repository already exist or does anyone have any tips?"

Should GO capture the context of activity

Is there a facility to construct GO annotations which would capture the context in which a given gene product is associated with (e.g.) a given function? Thus, for example, instead of annotating protein X to oxygen transport [function] one might annotate:

<protein X in cell component Y> to oxygen transport [function']

For MGI, this information is captured for cell types, anatomy and target molecules, but it is not a GO-wide policy. MGI captures this information in structured notes. So, for example MGI curators can make an annotation for gene product X having some function in oxygen transport in a lung epithelial cell. The hope is that if X is also annotated to oxygen transporter activity in lung epithelial cell and plasma membrane of a lung epithelial cell, then we can draw a conclusion that X has the potential to execute a function of an oxygen transporter involved in oxygen transport in the plasma membrane of a lung epithelial cell. In the right spatio-temporal context, the function will be executed.


Encouraging community annotation

  1. How can we help community scientists to get credit for their GO work? Could we publish a short account of the work of individual content meetings with all experts having author credit? Could groups providing manual annotation of their own gene products of interest get a small publication giving an account of their contribution and of possible applications?

Annotation Tool

  1. Who is the target audience for the annotation tool?
  2. How does the idea of an annual functional annotation tool bake-off sound to you? How would you approach it? Can it be used to keep orphaned genomes up-to-date?
  3. Almost everyone agrees that the GO Consortium should develop an annotation tool, [but not all people since most groups do annotation within the context of their genome annotation stream] Once we begin asking questions about what the tool should do, it becomes clear that every group has very different ideas about what such a tool should do. Groups within the GO tend to advocate for the development of a tool that satisfies their own research efforts at the moment, because that is the immediate need as they understand it. Could the advisors suggest a way of determining which of these special-purpose applications are best for the GO user community as a whole, so that we can decide how to focus our development efforts? What's the best way to encourage inter-group communication to foster collaboration on tools and minimize redundant development efforts? How can we decide whether specific, highly targeted annotation tools or general, lowest-common-denominator annotation tools would be most useful to our user community?

Agenda

In order to meet our objectives, we need to present all of the material in the morning. The presentations will focus on the goals and progress of the Consortium and will reflect the new organization and management of the GOC developed through the process of writing the GO Consortium competitive renewal. The times noted here are expected to be divided between a short presentation and then focused discussion on the topic.


    • - 8:00AM: Introductions (picture on display and circle everyone)
    • - 8:10 AM: Overview of Grant Aims (Michael-20 minutes)
    • - 8:30 AM: Reference Genomes (Rex - 30 minutes)
    • - 9:00 AM: Ontology/Database Developments - Biological (Midori and David-30 minutes)
    • - 9:30 AM: Ontology Developments – Technical (Chris-30 minutes)
    • -10:00 AM: Break (15 min)
    • -10:15 AM: Emerging Genomes (Jen-20 minutes)
    • -10:35AM: User Advocacy including AmiGO (Eurie & Jane- 20 minutes)
    • -10:55AM: New management structure (Suzi- 20 minutes)
    • -11:15AM: Summary of issues we face: pink sheets and more (Judy- 20 minutes)
    • -11:35AM: Open Discussion (all)
    • -12:30PM – 1:30PM, working lunch for Advisors and GO-top
    • - 1:30PM - 3:00PM, Advisors powwow
    • - 3:00 PM - break
    • - 3:30PM - 5:00PM, feedback from Advisors
    • - 6:00PM dinner with everyone

Participants

  • GO representatives
  1. Michael Ashburner
  2. Judy Blake
  3. Eurie Hong
  4. Suzi Lewis
  5. Rex Chisholm
  6. Jen Clark
  7. Midori Harris
  8. David Hill
  9. Ben Hitz (representing production services)
  10. Jane Lomax
  11. Chris Mungall
  12. Simon Twigger
  13. Tanya Berardini
  14. Nicky Mulder
  • GO Advisors
  1. Larry Hunter
  2. Lynette Hirschman
  3. Barry Smith
  4. David States
  5. Mike Tyers
  6. Craig Neville-Manning (unable to attend)
  7. Peter Tarczy-Harnach
  8. Ian Dix (Courtland Yockey attending)
  • NIH Representative
  1. Peter Good
  • Other possible attendees
  1. Monte Westerfield?

Venue

Accommodation: Silver Cloud Inns / Seattle-University Village

 5036 25th Avenue NE
 Seattle, WA 98105
 Phone: 206-526-5200

Slides