SLC GO Consortium Meeting: Difference between revisions

From GO Wiki
Jump to navigation Jump to search
mNo edit summary
 
(33 intermediate revisions by 7 users not shown)
Line 17: Line 17:
==='''Cost'''===
==='''Cost'''===
* $50.00; Receipt will be provided at registration. We will be taking cash and checks.  Checks should be made payable to the Department of Human Genetics. The address if needed for the check is: Eccles Institute of Human Genetics, Building 533, University of Utah, 15 North 2030 East, Salt Lake City, Utah. 84112.
* $50.00; Receipt will be provided at registration. We will be taking cash and checks.  Checks should be made payable to the Department of Human Genetics. The address if needed for the check is: Eccles Institute of Human Genetics, Building 533, University of Utah, 15 North 2030 East, Salt Lake City, Utah. 84112.
* Dinner: cost TBD; we are looking for a sponsor
* Dinner: cost TBD; No sponsor is provided.
* You are responsible for your lodging costs
* You are responsible for your lodging costs


Line 76: Line 76:
##[[Sensu Main Page|Sensu]]
##[[Sensu Main Page|Sensu]]
##Disjoint violations corrected
##Disjoint violations corrected
##[[Cardiovascular physiology/development|Cardio]] & [[Muscle Development|muscle]] content meeting work
##[[Cardiovascular physiology/development|Cardio]] & [[Muscle Biology|muscle]] content meeting work
##[[Electron transport]]
##[[Electron transport]]
##[[Progress_Reports_2008#Metrics|SourceForge statistics]]
##[[Progress_Reports_2008#Metrics|SourceForge statistics]]
Line 83: Line 83:
*break 10:45-11:00
*break 10:45-11:00


*11:00-noon Ontology Development continued
*11:00-noon Ontology Development continued#Work in progress
#Work in progress
##[[Function-Process Links]] (Harold, Chris, Jen)
##[[Function-Process Links]] (Harold, Chris, Jen)


Line 128: Line 127:


*3:30 - 4:00 Overall Project Management
*3:30 - 4:00 Overall Project Management
# Structure of Management Groups
## What is working
## What is working
## (and What is not) Identify communication bottlenecks and time-sinks, knowing who is responsible and time estimates
## (and What is not)  
#Identify communication bottlenecks and time-sinks, knowing who is responsible and time estimates
 


*4:00 - 4:30 Outstanding issues (continued)
*4:00 - 4:30 Outstanding issues (continued)
#[[Discuss]] how we are going to handle 'response to drug' SF 1242405 and 1494526 and 'response to toxin' SF1658374.  Are they normal biological processes? Response to X ([[Response to drug|drug]], toxin) Response to Chemical (see also [[Use of Response To Terms in Annotation]], for a related issue)(David and Tanya)
#[[Discuss]] Establishment and Maintenance Terms
#[[Discuss]]Establishment and Maintenance Terms
#contributes_to qualifier BAD-used to mean a number of different things (Val &Pascale)
#[[Discuss]]Chemical derivatives and metabolism terms
# [[Discuss]] IMP and the with column. Guidelines for what goes there - phenotype ID, genotype ID
# [[Discuss]] IMP and the with column. Guidelines for what goes there - phenotype ID, genotype ID
# [[Discuss]] how we are going to handle function terms such as activators and inhibitors. When do we create them and how do they relate to processes and other functions. (David and Tanya)
# [[Discuss]] how we are going to handle 'response to drug' SF 1242405 and 1494526 and 'response to toxin' SF1658374. Are they normal biological processes? Response to X ([[Response to drug|drug]], toxin) Response to Chemical (see also [[Use of Response To Terms in Annotation]], for a related issue)(David and Tanya)
# [[Discuss]] How should we handle derivative compounds in the metabolism parts of GO--should we continue to include metabolism of X derivatives as is_a descendants of X metabolism? It can be confusing. For example, SFs [https://sourceforge.net/tracker/index.php?func=detail&aid=1885151&group_id=36855&atid=440764 1885151] and [https://sourceforge.net/tracker/index.php?func=detail&aid=1847808&group_id=36855&atid=440764 1847808] note that 'gamma-aminobutyric acid metabolic process' is a descendant of 'fatty acid metabolic process' because GABA is a derivative of butyric acid. (Midori and Val)
# [[Discuss]] Chemical derivatives and metabolism terms: How should we handle derivative compounds in the metabolism parts of GO--should we continue to include metabolism of X derivatives as is_a descendants of X metabolism? It can be confusing. For example, SFs [https://sourceforge.net/tracker/index.php?func=detail&aid=1885151&group_id=36855&atid=440764 1885151] and [https://sourceforge.net/tracker/index.php?func=detail&aid=1847808&group_id=36855&atid=440764 1847808] note that 'gamma-aminobutyric acid metabolic process' is a descendant of 'fatty acid metabolic process' because GABA is a derivative of butyric acid. (Midori and Val)




Line 143: Line 144:
# Collaboration with other projects
# Collaboration with other projects
## PRO: Protein Ontology (Harold)
## PRO: Protein Ontology (Harold)
## Reactome
## Reactome (Peter)
## Panther
## Panther (Paul)
## NCBO
## NCBO (Suzi)
## MP-PATO
## MP-PATO (Chris)
## CARO
## CARO (Michael)
 
# At one of our meetings Chris suggested we might keep an archive of gene sets used in publications that were analyzed using GO. Should we do this? (David)
 
# Review Action Items from Princeton meeting [http://wiki.geneontology.org/index.php/GO_18th_Consortium_Meeting_Minutes_Day_1#Summary_of_action_Items_from_Day_1][http://wiki.geneontology.org/index.php/GO_18th_Consortium_Meeting_Minutes_Day_2#Summary_of_Action_points_from_Day_2] and from Cambridge
 
* 5:00 Take Group Picture


# Review Action Items from Princeton
* 6:30 Dinner
meeting [http://wiki.geneontology.org/index.php/GO_18th_Consortium_Meeting_Minutes_Day_1#Summary_of_action_Items_from_Day_1][http://wiki.geneontology.org/index.php/GO_18th_Consortium_Meeting_Minutes_Day_2#Summary_of_Action_points_from_Day_2] and from Cambridge






<b>Wed AM</b>
<b>Wed AM</b>
 
# [[Discuss]] Consistent use of identifier prefixes - GO.xrf_abbs and col1 and 2 of GAF. MGI and RGD problem. See [[Identifiers]]
# Obligatory Evidence Code discussion
# [[Discuss]] contributes_to qualifier BAD-used to mean a number of different things (Val)
## [[Discuss]]should we come up with a long term plan for abandoning evidence "codes" and use ECO IDs instead, with the benefits of ontology identifier lifecycle management (Chris)
# Report results of experiment on finding annotation errors using GO-taxon links. (Jen)
##[[Discuss]]process for editing ECO: curators, trackers
# Obligatory Evidence Code [[discussion]]
## [[Discuss]] of evidence codes, specifically the need for a distinction between curated and non-curated annotations.  For all of our current evidence codes there can be either curated or not.  A HTP protein array study's results of substrate binding would be IDA with only a few of these results being reviewed by a curator.  Those results would still have the IDA code.  There are examples of this for all the evidence codes but one -IEA.  A proposal from Kara,  "<b>Remove IEA as an evidence code.</b> Create a new property for GO annotations (or add a new type of qualifier) that captures how the annotation was done:  manual or automated.  Everything that is currently IEA would be given the 'automated' property/qualifier, and then would be given a new vidence code as appropriate (mostly a flavor of ISS I would assume). There can be a rule that all 'automated' annotations that are a flavor of ISS must have a 'with' value. This would allow us to use 'RCA' as appropriate, in some cases they'd be 'manual', in others, they'd be 'automated'.  I realize the issues involved in making such a drastic change, so I understand if we don't go there, but I do think that some approach such as the one above is the best representation of the information that we are trying to capture." (Mike; Kara will not be attending)
## ISS/IEA evidence code usage for annotations made using orthology sets (Emily)
meeting[http://wiki.geneontology.org/index.php/Outstanding_Action_Items_from_17th_GOC_Meeting%2C_Cambridge_UK]
## Should we come up with a long term plan for abandoning evidence "codes" and use ECO IDs instead, with the benefits of ontology identifier lifecycle management (Chris)
## process for editing ECO: curators, trackers
## Discussion specifically of the need for a distinction between curated and non-curated annotations.  For all of our current evidence codes there can be either curated or not.  A HTP protein array study's results of substrate binding would be IDA with only a few of these results being reviewed by a curator.  Those results would still have the IDA code.  There are examples of this for all the evidence codes but one -IEA.  A proposal from Kara,  "<b>Remove IEA as an evidence code.</b> Create a new property for GO annotations (or add a new type of qualifier) that captures how the annotation was done:  manual or automated.  Everything that is currently IEA would be given the 'automated' property/qualifier, and then would be given a new vidence code as appropriate (mostly a flavor of ISS I would assume). There can be a rule that all 'automated' annotations that are a flavor of ISS must have a 'with' value. This would allow us to use 'RCA' as appropriate, in some cases they'd be 'manual', in others, they'd be 'automated'.  I realize the issues involved in making such a drastic change, so I understand if we don't go there, but I do think that some approach such as the one above is the best representation of the information that we are trying to capture." (Mike; Kara will not be attending) meeting
[http://wiki.geneontology.org/index.php/Outstanding_Action_Items_from_17th_GOC_Meeting%2C_Cambridge_UK]


# Summarize Action Items from this meeting
# Summarize Action Items from this meeting
Line 168: Line 177:
At the moment this is a free-form list of items suggested for discussion.
At the moment this is a free-form list of items suggested for discussion.


# [[Discuss]] IMP and the with column. Guidelines for what goes there - phenotype ID, genotype ID
# [[not on agenda]] At one of our meetings Chris suggested we might keep an archive of gene sets used in publications that were analyzed using GO. Should we do this? (David)
# [[Discuss]] how we are going to handle function terms such as activators and inhibitors. When do we create them and how do they relate to processes and other functions. (David and Tanya)
# [[DISCUSS ]] contributes_to qualifier BAD-used to mean a number of different things (Val)
# [[Discuss]] How should we handle derivative compounds in the metabolism parts of GO--should we continue to include metabolism of X derivatives as is_a descendants of X metabolism? It can be confusing. For example, SFs [https://sourceforge.net/tracker/index.php?func=detail&aid=1885151&group_id=36855&atid=440764 1885151] and [https://sourceforge.net/tracker/index.php?func=detail&aid=1847808&group_id=36855&atid=440764 1847808] note that 'gamma-aminobutyric acid metabolic process' is a descendant of 'fatty acid metabolic process' because GABA is a derivative of butyric acid. (Midori and Val)
# [[DISCUSS]] Evidence Codes
# Discuss how we are going to handle 'response to drug' SF 1242405 and 1494526 and 'response to toxin' SF1658374.  Are they normal biological processes? (David and Tanya)
## process for editing ECO: curators, trackers
# Discuss function-process links. What relationships should we use? Who will make the assertions?(Harold, Jen & Chris)
## should we come up with a long term plan for abandoning evidence "codes" and use ECO IDs instead, with the benefits of ontology identifier lifecycle management (Chris)
## See current work and discussions [[Function-Process_Links]]
## Discussion of evidence codes, specifically the need for a distinction between curated and non-curated annotations.  For all of our current evidence codes there can be either curated or not.  A HTP protein array study's results of substrate binding would be IDA with only a few of these results being reviewed by a curator.  Those results would still have the IDA code.  There are examples of this for all the evidence codes but one -IEA.  A proposal from Kara,  "<b>Remove IEA as an evidence code.</b> Create a new property for GO annotations (or add a new type of qualifier) that captures how the annotation was done:  manual or automated.  Everything that is currently IEA would be given the 'automated' property/qualifier, and then would be given a new vidence code as appropriate (mostly a flavor of ISS I would assume). There can be a rule that all 'automated' annotations that are a flavor of ISS must have a 'with' value. This would allow us to use 'RCA' as appropriate, in some cases they'd be 'manual', in others, they'd be 'automated'.  I realize the issues involved in making such a drastic change, so I understand if we don't go there, but I do think that some approach such as the one above is the best representation of the information that we are trying to capture." (Mike; Kara will not be attending)
# [[DISCUSS]] Consistent use of identifier prefixes - GO.xrf_abbs and col1 and 2 of GAF. MGI and RGD problem. See [[Identifiers]]
# [[DONE]] Discuss function-process links. What relationships should we use? Who will make the assertions?(Harold, Jen & Chris)
## [[DONE]] See current work and discussions [[Function-Process_Links]]
# [[DONE]] how we are going to handle function terms such as activators and inhibitors. When do we create them and how do they relate to processes and other functions. (David and Tanya)
# [[DONE]] Have Tanya give a presentation on implementation of 'regulates' and the work that she and David and Chris (and others) have done in this area.  'Regulates' will go live at the end of March, so there may be discussion points on this. We will also discuss 'regulation of biological quality' and how their negative and positive children relate.
# [[DONE]] IMP and the with column. Guidelines for what goes there - phenotype ID, genotype ID
# [[DONE]] How should we handle derivative compounds in the metabolism parts of GO--should we continue to include metabolism of X derivatives as is_a descendants of X metabolism? It can be confusing. For example, SFs [https://sourceforge.net/tracker/index.php?func=detail&aid=1885151&group_id=36855&atid=440764 1885151] and [https://sourceforge.net/tracker/index.php?func=detail&aid=1847808&group_id=36855&atid=440764 1847808] note that 'gamma-aminobutyric acid metabolic process' is a descendant of 'fatty acid metabolic process' because GABA is a derivative of butyric acid. (Midori and Val)
# [[DONE]] how we are going to handle 'response to drug' SF 1242405 and 1494526 and 'response to toxin' SF1658374.  Are they normal biological processes? (David and Tanya)
# (from ref genomes group): Would it be useful to add the 'comprehensively annotated' tag to all genes (not just the ref genome genes)?  Either in the gene association file or in the database somehow
# (from ref genomes group): Would it be useful to add the 'comprehensively annotated' tag to all genes (not just the ref genome genes)?  Either in the gene association file or in the database somehow
# At one of our meetings Chris suggested we might keep an archive of gene sets used in publications that were analyzed using GO. Should we do this? (David)
# [[DONE]]  Annotation Objects: how to handle GO annotation to specific isoforms, modified proteins, etc. (Harold and Chris) [[http://gocwiki.geneontology.org/index.php/Annotation_of_Alternate_Spliceforms]])
# Collaboration with other projects
# [[DONE]] Proteases: any objections to making a lot of protease activity MF terms obsolete, on the grounds that they're gene products? See [[Proteases]] for a detailed proposal. (Midori et al.)
## PRO: Protein Ontology (Harold)
# [[Deferred by Peter ]] Granularity in GO - Peter (Reactome) has raised this question again, because we've discussed it but not fully reached consensus. An example from Reactome's perspective: "GO, for instance, has a single term GO:0031261 'DNA replication preinitiation complex', with no child terms, while Reactome enumerates all the intermediate stages of its assembly during G1 of the cell cycle and activation at G1/S." Should GO simply add the very specific terms from Reactome? (I think the same question applies to GO process terms -- many Reactome events are much more fine-grained than most GO processes.) (Midori and Peter)
## Reactome
# [[DONE]]Collaboration with other projects
## Panther
## PRO: Protein Ontology (Harold) - ok
## NCBO
## Reactome - ok
## MP-PATO
## Panther - ok
## CARO
## NCBO -ok
# [[DONE]] Annotation Objects: how to handle GO annotation to specific isoforms, modified proteins, etc. (Harold and Chris [[http://gocwiki.geneontology.org/index.php/Annotation_of_Alternate_Spliceforms]])
## MP-PATO-ok
# Report results of experiment on finding annotation errors using GO-taxon links. (Jen)
## CARO - ok
#[[ DONE ]] Report results of experiment on finding annotation errors using GO-taxon links. (Jen)
# Should we approach authors of papers describing computationally assisted annotation of genes from non-MOD species with GO terms to integrate their results into the GO database?  If not, can we take any advantage of these studies? For example, click on B2G in Papers on this site [[http://blast2go.de/]] (Tanya)


# Should we approach authors of papers describing computationally assisted annotation of genes from non-MOD species with GO terms to integrate their results into the GO database?  If not, can we take any advantage of these studies? For example, click on B2G in Papers on this site [[http://blast2go.de/]]  (Tanya)
# Proteases: any objections to making a lot of protease activity MF terms obsolete, on the grounds that they're gene products? See [[Proteases]] for a detailed proposal. (Midori et al.)
# Granularity in GO - Peter (Reactome) has raised this question again, because we've discussed it but not fully reached consensus. An example from Reactome's perspective: "GO, for instance, has a single term GO:0031261 'DNA replication preinitiation complex', with no child terms, while Reactome enumerates all the intermediate stages of its assembly during G1 of the cell cycle and activation at G1/S." Should GO simply add the very specific terms from Reactome? (I think the same question applies to GO process terms -- many Reactome events are much more fine-grained than most GO processes.) (Midori and Peter)
# contributes_to qualifier BAD-used to mean a number of different things (Val)
# process for editing ECO: curators, trackers
# should we come up with a long term plan for abandoning evidence "codes" and use ECO IDs instead, with the benefits of ontology identifier lifecycle management (Chris)
# Discussion of evidence codes, specifically the need for a distinction between curated and non-curated annotations.  For all of our current evidence codes there can be either curated or not.  A HTP protein array study's results of substrate binding would be IDA with only a few of these results being reviewed by a curator.  Those results would still have the IDA code.  There are examples of this for all the evidence codes but one -IEA.  A proposal from Kara,  "<b>Remove IEA as an evidence code.</b> Create a new property for GO annotations (or add a new type of qualifier) that captures how the annotation was done:  manual or automated.  Everything that is currently IEA would be given the 'automated' property/qualifier, and then would be given a new vidence code as appropriate (mostly a flavor of ISS I would assume). There can be a rule that all 'automated' annotations that are a flavor of ISS must have a 'with' value. This would allow us to use 'RCA' as appropriate, in some cases they'd be 'manual', in others, they'd be 'automated'.  I realize the issues involved in making such a drastic change, so I understand if we don't go there, but I do think that some approach such as the one above is the best representation of the information that we are trying to capture." (Mike; Kara will not be attending)
# Consistent use of identifier prefixes - GO.xrf_abbs and col1 and 2 of GAF. MGI and RGD problem. See [[Identifiers]]


# [[DONE]] Have Tanya give a presentation on implementation of 'regulates' and the work that she and David and Chris (and others) have done in this area.  'Regulates' will go live at the end of March, so there may be discussion points on this. We will also discuss 'regulation of biological quality' and how their negative and positive children relate.
[[Category:GO Consortium Meetings‏‎‏‎]]

Latest revision as of 10:06, 15 April 2019

Venue

Meeting Room

  • Conference Room 4 is the large conference room on the 4th floor, located behind the front desk to the right.
  • We are not allowed to put posters up to direct you, but if you get lost, the library staff will point you in the right direction.

Dates:

  • April 22-23, 2008
  • We will be providing lunch on the 22nd
  • The meeting ends at lunchtime on the 23rd.
  • Dinner will be held on Tuesday 22nd April at Cannellas. The reservation is for 6.30, and it is located one block South from the library. The entrees are spagetti, grilled portobella stack, chicken marsala and grilled salmon.

Cost

  • $50.00; Receipt will be provided at registration. We will be taking cash and checks. Checks should be made payable to the Department of Human Genetics. The address if needed for the check is: Eccles Institute of Human Genetics, Building 533, University of Utah, 15 North 2030 East, Salt Lake City, Utah. 84112.
  • Dinner: cost TBD; No sponsor is provided.
  • You are responsible for your lodging costs

Lodging:

  • The University GuestHouseis a good place to stay as it is a short walk to HSEB. If you are also attending the Reference Genome Meeting, it is a small walk from the that venue, and a TRAX ride from the library - door to door. The one way fare is $1.75 or a day pass can be purchased for $4.50. Tickets can be purchased at the platform. A map is available here: http://www.rideuta.com/images/TRAX082607.jpg

TRAX schedule from the guest house to the library. (The library opens at 9 am.)

MEDICAL CENTER FT DOUGLAS SOUTH CAMPUS STADIUM TRAX STATION 900 EAST & 400 SOUTH (TRAX) TROLLEY TRAX STATION LIBRARY TRAX STATION
08:23 AM 08:25 AM 08:27 AM 08:29 AM 08:32 AM 08:34 AM 08:36 AM
08:38 AM 08:40 AM 08:42 AM 08:44 AM 08:47 AM 08:49 AM 08:51 AM


  • Please make reservations early as it is a popular place and there will be extra people in town for the marathon.
  • If you need to make other arrangements let me know (Karen E) and I will find a different hotel.
  • The | University_Marriott at the university is a cab ride from both venues.
  • There are many hotels downtown with reasonable rates such as the Little America and the Red Lion.
  • The Downtown_Marriott within walking distance of the library, but it is $259 per night!
  • To book a taxi from the library to the airport: 801 521 2100

Registration

Please [Register] for one or both meetings.

Agenda

Tue am -

  • 9:00- Intro, review agenda
  • 9:15 - 10:00 - Ref Genome report (Pascale)
  1. Annotation Metrics
  2. Annotation Pipeline
  3. Annotation to Isoforms (1 new column)
  • 10:00-10:45 Ontology Development
  1. Report on accomplishments (15 min. for 1st several items; 30 min. for 'regulates')
    1. Sensu
    2. Disjoint violations corrected
    3. Cardio & muscle content meeting work
    4. Electron transport
    5. SourceForge statistics
    6. Regulates (including activators and inhibitors)
  • break 10:45-11:00
  • 11:00-noon Ontology Development continued#Work in progress
    1. Function-Process Links (Harold, Chris, Jen)


      1. Considerations on Glycolysis [[1]] and TCA cycle [[2]]
      2. Also see "term granularity" below
    1. Cross products
      1. Regulation cross-products
      2. BP-xp-CC
      3. BP-xp-CL
      4. Cross Product Timeline
      5. AmiGO_1_6 xps in AmiGO
    2. Taxon
    3. Lung Development
  1. Future work
    1. Signal transduction
    2. Morphogenesis
    3. Transport
    4. Metabolism cleanup
  2. Outstanding issues (round one)
    1. Proteases
  • noon - lunch

Tues Afternoon Session -

  • 1:30 - 2:30
  1. Advocacy/Outreach/Collaborations (Jen/Jane)
    1. Help Desk (stats report) and newsletter
    2. User Oversight of AmiGO functionality (look and feel)
    3. Active outreach collaborations
    4. New outreach contacts.
  2. AmiGO, GO Database
    1. Software group progress report
    2. ReferenceGenes - DB management extensions
  3. SO (Karen E)
  • break
  • 3:15 - 3:30 Software (continued),
  1. OBO-Edit
    1. Might include demo of Michael Schroeder's plug-in
  • 3:30 - 4:00 Overall Project Management
  1. Structure of Management Groups
    1. What is working
    2. (and What is not)
  2. Identify communication bottlenecks and time-sinks, knowing who is responsible and time estimates


  • 4:00 - 4:30 Outstanding issues (continued)
  1. Discuss Establishment and Maintenance Terms
  2. contributes_to qualifier BAD-used to mean a number of different things (Val &Pascale)
  3. Discuss IMP and the with column. Guidelines for what goes there - phenotype ID, genotype ID
  4. Discuss how we are going to handle 'response to drug' SF 1242405 and 1494526 and 'response to toxin' SF1658374. Are they normal biological processes? Response to X (drug, toxin) Response to Chemical (see also Use of Response To Terms in Annotation, for a related issue)(David and Tanya)
  5. Discuss Chemical derivatives and metabolism terms: How should we handle derivative compounds in the metabolism parts of GO--should we continue to include metabolism of X derivatives as is_a descendants of X metabolism? It can be confusing. For example, SFs 1885151 and 1847808 note that 'gamma-aminobutyric acid metabolic process' is a descendant of 'fatty acid metabolic process' because GABA is a derivative of butyric acid. (Midori and Val)


  • 4:30 - 5:00
  1. Collaboration with other projects
    1. PRO: Protein Ontology (Harold)
    2. Reactome (Peter)
    3. Panther (Paul)
    4. NCBO (Suzi)
    5. MP-PATO (Chris)
    6. CARO (Michael)
  1. At one of our meetings Chris suggested we might keep an archive of gene sets used in publications that were analyzed using GO. Should we do this? (David)
  1. Review Action Items from Princeton meeting [3][4] and from Cambridge
  • 5:00 Take Group Picture
  • 6:30 Dinner


Wed AM

  1. Discuss Consistent use of identifier prefixes - GO.xrf_abbs and col1 and 2 of GAF. MGI and RGD problem. See Identifiers
  2. Discuss contributes_to qualifier BAD-used to mean a number of different things (Val)
  3. Report results of experiment on finding annotation errors using GO-taxon links. (Jen)
  4. Obligatory Evidence Code discussion
    1. ISS/IEA evidence code usage for annotations made using orthology sets (Emily)
    2. Should we come up with a long term plan for abandoning evidence "codes" and use ECO IDs instead, with the benefits of ontology identifier lifecycle management (Chris)
    3. process for editing ECO: curators, trackers
    4. Discussion specifically of the need for a distinction between curated and non-curated annotations. For all of our current evidence codes there can be either curated or not. A HTP protein array study's results of substrate binding would be IDA with only a few of these results being reviewed by a curator. Those results would still have the IDA code. There are examples of this for all the evidence codes but one -IEA. A proposal from Kara, "Remove IEA as an evidence code. Create a new property for GO annotations (or add a new type of qualifier) that captures how the annotation was done: manual or automated. Everything that is currently IEA would be given the 'automated' property/qualifier, and then would be given a new vidence code as appropriate (mostly a flavor of ISS I would assume). There can be a rule that all 'automated' annotations that are a flavor of ISS must have a 'with' value. This would allow us to use 'RCA' as appropriate, in some cases they'd be 'manual', in others, they'd be 'automated'. I realize the issues involved in making such a drastic change, so I understand if we don't go there, but I do think that some approach such as the one above is the best representation of the information that we are trying to capture." (Mike; Kara will not be attending) meeting
[5]
  1. Summarize Action Items from this meeting

Agenda Items

At the moment this is a free-form list of items suggested for discussion.

  1. not on agenda At one of our meetings Chris suggested we might keep an archive of gene sets used in publications that were analyzed using GO. Should we do this? (David)
  2. DISCUSS contributes_to qualifier BAD-used to mean a number of different things (Val)
  3. DISCUSS Evidence Codes
    1. process for editing ECO: curators, trackers
    2. should we come up with a long term plan for abandoning evidence "codes" and use ECO IDs instead, with the benefits of ontology identifier lifecycle management (Chris)
    3. Discussion of evidence codes, specifically the need for a distinction between curated and non-curated annotations. For all of our current evidence codes there can be either curated or not. A HTP protein array study's results of substrate binding would be IDA with only a few of these results being reviewed by a curator. Those results would still have the IDA code. There are examples of this for all the evidence codes but one -IEA. A proposal from Kara, "Remove IEA as an evidence code. Create a new property for GO annotations (or add a new type of qualifier) that captures how the annotation was done: manual or automated. Everything that is currently IEA would be given the 'automated' property/qualifier, and then would be given a new vidence code as appropriate (mostly a flavor of ISS I would assume). There can be a rule that all 'automated' annotations that are a flavor of ISS must have a 'with' value. This would allow us to use 'RCA' as appropriate, in some cases they'd be 'manual', in others, they'd be 'automated'. I realize the issues involved in making such a drastic change, so I understand if we don't go there, but I do think that some approach such as the one above is the best representation of the information that we are trying to capture." (Mike; Kara will not be attending)
  4. DISCUSS Consistent use of identifier prefixes - GO.xrf_abbs and col1 and 2 of GAF. MGI and RGD problem. See Identifiers
  5. DONE Discuss function-process links. What relationships should we use? Who will make the assertions?(Harold, Jen & Chris)
    1. DONE See current work and discussions Function-Process_Links
  6. DONE how we are going to handle function terms such as activators and inhibitors. When do we create them and how do they relate to processes and other functions. (David and Tanya)
  7. DONE Have Tanya give a presentation on implementation of 'regulates' and the work that she and David and Chris (and others) have done in this area. 'Regulates' will go live at the end of March, so there may be discussion points on this. We will also discuss 'regulation of biological quality' and how their negative and positive children relate.
  8. DONE IMP and the with column. Guidelines for what goes there - phenotype ID, genotype ID
  9. DONE How should we handle derivative compounds in the metabolism parts of GO--should we continue to include metabolism of X derivatives as is_a descendants of X metabolism? It can be confusing. For example, SFs 1885151 and 1847808 note that 'gamma-aminobutyric acid metabolic process' is a descendant of 'fatty acid metabolic process' because GABA is a derivative of butyric acid. (Midori and Val)
  10. DONE how we are going to handle 'response to drug' SF 1242405 and 1494526 and 'response to toxin' SF1658374. Are they normal biological processes? (David and Tanya)
  11. (from ref genomes group): Would it be useful to add the 'comprehensively annotated' tag to all genes (not just the ref genome genes)? Either in the gene association file or in the database somehow
  12. DONE Annotation Objects: how to handle GO annotation to specific isoforms, modified proteins, etc. (Harold and Chris) [[6]])
  13. DONE Proteases: any objections to making a lot of protease activity MF terms obsolete, on the grounds that they're gene products? See Proteases for a detailed proposal. (Midori et al.)
  14. Deferred by Peter Granularity in GO - Peter (Reactome) has raised this question again, because we've discussed it but not fully reached consensus. An example from Reactome's perspective: "GO, for instance, has a single term GO:0031261 'DNA replication preinitiation complex', with no child terms, while Reactome enumerates all the intermediate stages of its assembly during G1 of the cell cycle and activation at G1/S." Should GO simply add the very specific terms from Reactome? (I think the same question applies to GO process terms -- many Reactome events are much more fine-grained than most GO processes.) (Midori and Peter)
  15. DONECollaboration with other projects
    1. PRO: Protein Ontology (Harold) - ok
    2. Reactome - ok
    3. Panther - ok
    4. NCBO -ok
    5. MP-PATO-ok
    6. CARO - ok
  16. DONE Report results of experiment on finding annotation errors using GO-taxon links. (Jen)
  17. Should we approach authors of papers describing computationally assisted annotation of genes from non-MOD species with GO terms to integrate their results into the GO database? If not, can we take any advantage of these studies? For example, click on B2G in Papers on this site [[7]] (Tanya)