Signaling Workshop February 2011: Difference between revisions

From GO Wiki
Jump to navigation Jump to search
Line 334: Line 334:
== Minutes ==  
== Minutes ==  
(preliminary version)
(preliminary version)
SO-Sandra Orchard; VW-Val Wood; AT-Andrea Townsend-Nicholson; RL-Ruth Lovering; RF- Rebecca Foulger; PD-Peter D’Eustachio; MH-Midori Harris; PR-Paola Roncaglia; VK-Varsha Khodiyar; ST-Susan Tweedie; AC-Andrew Chat-ra, SJ-Steve Jupe


[[Wed morning session]]
[[Wed morning session]]
Line 346: Line 349:
''Discussions''
''Discussions''
*Because 'signaling' is so broad, and includes multi-organism signaling, nerve transmission etc, it's hard to find a common start point.
*Because 'signaling' is so broad, and includes multi-organism signaling, nerve transmission etc, it's hard to find a common start point.
*For signaling between cells, its important to know which cell the signal comes from (we already have terms such as mesoderm-endoderm cell signaling). We also need to distinguish 'signaling' from 'signal transduction', which starts with a receptor receiving a signal (E.g. ligand-receptor binding).
*We need to distinguish 'signaling' from 'signal transduction', which starts with a receptor receiving a signal (E.g. ligand-receptor binding).
*SO: There's overlap between the end of one pathway and the start of the next if you start with 'ligand production'.
*For signaling between cells, its important to know which cell the signal comes from (we already have terms such as mesoderm-endoderm cell signaling).  
*For some pathways, the signal is synthesised in response to a stimulus. In some pathways secretion is the regulatory point (eg insulin receptor signaling where insulin is made in advance and secreted upon glucose stimulus). For some pathways, the stimulus is the signal (eg nutrient sensing).
*For some pathways, the signal is synthesised in response to a stimulus. In some pathways secretion is the regulatory point (eg insulin receptor signaling where insulin is made in advance and secreted upon glucose stimulus). For some pathways, the stimulus is the signal (eg nutrient sensing).


Line 354: Line 359:
*''Decision'': transport of signal between cells is included in signaling (this fits with  terms such as endocrine signaling, paracrine signaling etc).
*''Decision'': transport of signal between cells is included in signaling (this fits with  terms such as endocrine signaling, paracrine signaling etc).
*''Decision'': 'regulation of signaling' does not make much sense as a term. Regulation of 'signal transduction' may be better. What we mean is 'regulation of a signaling process'.
*''Decision'': 'regulation of signaling' does not make much sense as a term. Regulation of 'signal transduction' may be better. What we mean is 'regulation of a signaling process'.
*'Decision'': AT: GO needs a 'glossary' of terms for things like 'signal' so we can standardise each definition. This should go on the signaling wiki.


''Actions''
''Actions''
Line 364: Line 370:
'''Endocrine vs Paracrine''' slide 8
'''Endocrine vs Paracrine''' slide 8


Should we distinguish between endocrine and paracrine signaling in GO and add these as new terms (paracrine signaling is currently a narrow synonym of cell-cell signaling)?  Endocrinologist reaction that Stan L contacted was mixed, some said fine to split, some said not possible to tell experimentally.
''Questions''
*Q: Should we distinguish between endocrine and paracrine signaling in GO and add these as new terms (paracrine signaling is currently a narrow synonym of cell-cell signaling)?   
 
 
'Discussions''
* Endocrinologist reaction that Stan L contacted was mixed, some said fine to split, some said not possible to tell experimentally.
* AT – can you really distinguish between auto and para?  These types of signaling are shown in intact organisms, not in cell based assays.


*''Decision:'' yes is it worth adding these but we should also add a comment that they should be used with care.
*''Decision:'' yes is it worth adding these but we should also add a comment that they should be used with care.
Line 370: Line 382:
*''AI:'' make new terms for endocrine signaling and paracrine signaling: BECKY.
*''AI:'' make new terms for endocrine signaling and paracrine signaling: BECKY.


----
'''signal transduction, signal transducer activity and signaling pathways'''
''Questions''
*Q: Where does signal transduction start and end?
''Discussions''
*RF: A signaling pathway ends with regulation of a downstream cellular process.
*There was a lot of discussions about whether transcription factors (TFs) should be part of a signaling pathway. AT said they should. ST said 'foxo' should be returned in a search for members of the insulin receptor signaling pathway.
*Do we need the term 'signal transducer'? The general consensus is that this term could be obsoleted. We need to look at children though, and how this would fit with the proposal to change 'receptor signaling protein activity' to 'signal transducing protein kinase activity'.
''Decisions''
* It's ok for the ligand to be annotated to 'signal transduction'.
* Signal transduction begins with reception of a signal (E.g. ligand-receptor binding) and ends with regulation of a downstream process.
* TFs are part of a pathway (eg part of signal transduction) because they pass the signal onto downstream RNA polymerase and regulate transcription.
* Glycogen Synthase (GS) is NOT part of the insulin receptor signaling pathway because it is performing the act of glycogen synthesis. It is not passing the signal on.
*AI: Check with David and Karen that this idea fits with transcription overhaul.


----
----
Line 382: Line 414:


''Discussions''
''Discussions''
* The comment and definition are not consistent: The comments say that both ligands and receptors can be annotated to 'initiation of signal transduction', but for many (eg signal transduction by light), there is no ligand gene product to annotate.
* VW:termination of signal transduction is like saying the last step in a pathway x which is meaningless.
* Discussion of whether the signal is always extracellular (AT: says Oxford Dictionary def of signal transduction includes an extracellular ligand). VW: SIN doesn't start with an extracellular ligand.


Both of these terms were made during previous revisions by the signaling work group. Although they seemed sensible at the time, it is not clear where they begin or end, or what products would be annotated to them. The comments say that both ligands and receptors can be annotated to 'initiation of signal transduction', but for many (eg signal transduction by light), there is no ligand gene product to annotate.
* ''Decision:'' The general conclusion was that these terms aren't useful and it makes more sense to remove them than tighten up the definitions.  Only 6 annotations have been made with initiation of signal transduction and these are better captured using function terms e.g. 'signal induction by light' would be better captured with a term 'light activated receptor activity'. There are no annotations to 'termination of signal transduction'.  
VW:termination of signal transduction is like saying the last step in a pathway x which is meaningless.
 
''Decision:'' The general conclusion was that these terms aren't useful and it makes more sense to remove them than tighten up the definitions.  Only 6 annotations have been made with initiation of signal transduction and these are better captured using function terms e.g. 'signal induction by light' would be better captured with a term 'light activated receptor activity'. There are no annotations to 'termination of signal transduction'.  


*''AI'': obsolete initiation of signal transduction and termination of signal transduction.
*''AI'': obsolete initiation of signal transduction and termination of signal transduction.
Line 394: Line 426:
'''impact on linking pathways to response to terms''' (see slide 24)
'''impact on linking pathways to response to terms''' (see slide 24)


'''Questions'''
''Questions'


We want to link pathways to the appropriate 'response to' terms in GO e.g insulin receptor signaling pathway is_a cellular response to insulin stimulus.  
*Q: We want to link pathways to the appropriate 'response to' terms in GO e.g insulin receptor signaling pathway is_a cellular response to insulin stimulus. Now the the ligand is included in the signaling pathway is it a problem that insulin itself would be returned in a search for 'cellular response to insulin stimulus'?  
*Q: Now the the ligand is included in the signaling pathway is it a problem that insulin itself would be returned in a search for 'cellular response to insulin stimulus'?  


'''Discussions:'''
''Discussions:''
Members of the insulin receptor signaling pathway from receptor-downwards, should be returned in a search for 'response to insulin'. The only way to get around it would be to create child terms for every pathway, beginning from an active receptor. The only difference between this and the parent term would be the ligand. In practise this would be a huge task and confusing to annotators. It would also be problematic when connecting the pathways to other events: do you choose the whole pathway, or just from receptor downwards?
Members of the insulin receptor signaling pathway from receptor-downwards, should be returned in a search for 'response to insulin'. The only way to get around it would be to create child terms for every pathway, beginning from an active receptor. The only difference between this and the parent term would be the ligand. In practise this would be a huge task and confusing to annotators. It would also be problematic when connecting the pathways to other events: do you choose the whole pathway, or just from receptor downwards?


There wasn't complete agreement about this (AT: didn't want see insulin in this search and thought it could be misleading but SO: would be very surprised if it wasn't there) but it was agreed that having a few potentially unexpected ligands in these searches was a small price to pay for the value of retaining the link to the response to term. We also agreed that the language sound ugly but there wasn't much we could do about it - i.e. the response to term x sounds like it would exclude x itself.  
There wasn't complete agreement about this (AT: didn't want see insulin in this search and thought it could be misleading but SO: would be very surprised if it wasn't there) but it was agreed that having a few potentially unexpected ligands in these searches was a small price to pay for the value of retaining the link to the response to term. We also agreed that the language sound ugly (PD) but there wasn't much we could do about it.
 
''Decisions''
* We can live with insulin being returned in a search for 'response to insulin stimulus'.


----
----
Line 412: Line 446:
''Questions:''
''Questions:''
*Q: Is cascade an appropriate name for these sub-pathways?
*Q: Is cascade an appropriate name for these sub-pathways?
*Q: Does there have to be amplification for it to be a 'cascade'?


''Discussions:''
''Discussions:''
Line 417: Line 452:


''AI'': Canvas wider opinion for best name
''AI'': Canvas wider opinion for best name
----
Receptors that bind multiple ligands, and ligands that bind multiple receptors
Problem
*To have the activity of an EGFR in function, the GP has to bind EGF and pass the signal on. In process, we are not consistent about whether we describe an EGFR as a receptor that binds EGF OR as the gene product that's called EGFR and can bind to a number of physiological ligands.
* To create process-function links, we have to be consistent.
Questions
*Does the end response depend on the ligand or receptor? (AT says both)
Decisions
We came up with the proposed ontology structure:
        EGF-activated signaling pathway
--[isa]EGF-activated pathway by FGFR
--[isa]EGF-activated pathway by EGFR
EGFR signaling pathway
--[isa]EGF-activated pathway by EGFR
--[isa]FGF-activated pathway by EGFR
FGFR signaling pathway
--[isa]EGF-activated pathway by FGFR
--[isa]FGF-activated pathway by FGFR
*This allows you to go in via the receptor or the ligand, depending on the experiment.
*In process 'EGFR' is the gene product called EGFR (this could get a PRO:ID).
''Decision:''
* Rename 'epidermal growth factor receptor activity' to 'epidermal growth factor-binding receptor activity' (with a HAS_PART EGF binding relationship).
''AI'': Run this proposal past Chris Mungall.


== Notes ==
== Notes ==

Revision as of 07:39, 21 February 2011

Return to GOC Main Page

Return to Signaling Main Page [[1]]

Please suggest agenda items, relevant papers and other information relating to signaling GO terms that you would like discussed at this workshop.



Participants

GO Annotators Remote Participants GO Editors GO Developers Signaling Experts
Dr Ruth Lovering (BHF-UCL)
Dr Varsha Khodiyar (BHF-UCL)
Dr Sandra Orchard (Intact)
Dr Peter D'Eustachio (Reactome)
Dr Rachael Huntley(GOA)
Dr Val Wood (PombeBase)
Dr Susan Tweedie (FlyBase)
Dr Alex Diehl
(MGI/UB Medical School;
GO curator/GO editor)
Dr Rebecca Foulger (GO Editorial)
Dr Midori Harris (GO Editorial; TBC)
Dr Paola Roncaglia (GO Editorial from March 2011)
Chris Mungall (Berkeley)
(Thursday only)
Dr Andrea Townsend-Nicholson (UCL)
Dr Andrew Winter (University Edinburgh)
Dr Andrew Chatr-aryamontri (University Edinburgh)

Where and When

Date: Wednesday 16th February and Thursday 17th February, 2011.

Where: EMBL-EBI.

Wednesday: A2-33

Thursday: Courtyard room

  • Travel instructions to the EBI, The Wellcome Trust Genome Campus, Hinxton, Cambridgeshire CB10 1SD [2]
  • Please let EBI reception know you have arrived, and Rebecca will come and meet you.

Accommodation

The following participants have rooms reserved on site for the nights of Tuesday 15th and Wednesday 16th February:

  • Dr Andrew Winter
  • Dr Andrew Chatr-aryamontri
  • Dr Andrea Townsend-Nicholson
  • Dr Ruth Lovering

The following participants have rooms reserved on site for the nights of Wednesday 16th February:

  • Dr Varsha Khodiyar

The following participants have rooms reserved on site for the nights of Monday 14th, Tuesday 15th, Wednesday 16th and Thursday 17th February.

  • Dr Paola Roncaglia

All other participants are local or have arranged their own accommodation. Please let Rebecca know if this changes.

Dinners

  • For participants arriving on the Tuesday, and for anyone else who would like to join us, a meal will be arranged on Tuesday 15th February at the Hinxton Red Lion, 7pm.

The Red Lion Menu's are available here: Please send Rebecca your choices by Friday 4th February.

Bar Alacarte puddings

  • A workshop meal will be arranged for Wednesday 16th February at the John Barleycorn in Duxford at 7pm. Menu is here
  • Lunches on the Wednesday and Thursday will be at the on-site Murrays restaurant. Lunch vouchers will be provided for all participants.
  • Tea and Coffee will be provided in the meeting rooms

FUNDING

The GO signaling workshop is jointly funded by GO (75%) and UCL-BHF (25%).

The EBI Industry Programme, lead by Dominic Clark, is kindly funding the lunches and the workshop dinner on the Wednesday evening.


Programme

Wednesday 16th February

  • Morning Minutes: Varsha Khodiyar
  • Afternoon Minutes: Susan Tweedie


9:00am Tea and Coffee
9:30am Welcome and introduction Rebecca
9:35am Introduction to GO Rebecca
9:50am Definition and Scope of signaling
Definition of 'Signal Generation'
Definition of 'Signal Transduction', including 'initiation' and 'termination'
11am Coffee break
11:20am Signaling pathways- how to define and group
Non-1:1 receptor-ligand combinations
12:30pm Lunch @ on-site campus restaurant
1:30pm The function of being a receptor
Nuclear hormone receptors
Defining Intracellular 'cassettes'
3:00 pm Coffee break
Connecting Intracellular 'cassettes' with upstream receptors
Receptor-mediated endocytosis and signaling
Transport of the ligand to the receiving cell- endocrine/paracrine signaling
Calcium-mediated signaling
5:30pm Close


Thursday 17th February

  • Morning Minutes: Rachael Huntley
  • Afternoon Minutes: Susan Tweedie


9:00am Tea and Coffee
9:30am Recap of decisions made on Day 1 Sandra
Oustanding/Unresolved issues from Wednesday: Receptor function (incl. P2X). RME. Sandra and Ruth
10am Chris will present work he's doing making automatic links between Reactome and GO signaling terms Chris Mungall
10:20am Run 'EGF-activated signaling pathway via "FGFR" proposal' past Chris. Ruth, Sandra and Chris
10:30am Work through the insulin receptor signaling pathway to create a curation manual for annotators Sandra and Ruth
Ligand-Receptor binding
Proteins downstream of ligand-receptor binding: Signaling via IRS1. Signaling via SHC/MAPK.
11am Coffee break
Pathway regulators
Proteins upstream of ligand-receptor binding: Insulin synthesis and secretion. Insulin receptor synthesis and processing.
12:30pm Lunch @ on-site campus restaurant
1:30pm Proteins involved in receptor-mediated endocytosis of ligand-receptor
3:00 pm Close... Thank you all very much

A summary of the Insulin receptor signaling pathway is here and here (it's separated into two pdfs because of size restrictions for uploads)

A list of components of the Insulin receptor signaling pathway is here


Relevant papers and other resources

Biochemistry of Signal Tranduction and Regulation. Gerhard Krauss. ISBN:3-527-29241-1

Cell Signaliing. John T. Hancock. ISBN:0199264678

The Biochemistry of Cell Signalling. Ernst J. M. Helmreich. ISBN: 0198508204

Signal Transduction. Gompert et al. ISBN: 9787030182173. (recommended by Andrea: provides a history on signal transduction)


Topics

A presentation on all the signaling topics is here


  • How the signaling node should be split
    • cellular vs tissue vs organismal?
    • pathway vs process?
  • Defining and placing:
    • signal transduction
    • signal transmission
    • signaling pathway
  • Defining initiation of signal transduction:
      • Currently used to annotate both ligands and receptors- is this correct?
      • Which of its children do we want to keep?
  • Defining 'creation of a signal':
    • Where does the process start... transcription/translation or later?
  • Receptor processing:
    • Should receptor processing terms go under 'signaling' to be siblings with the 'creation of a signal' term?
    • OR should receptor processing terms go under 'regulation of signaling'?
  • Connecting the networks
    • How to connect the cell surface receptors with the intracellular 'modules' they signal through
  • Dealing with ligand-receptor signaling that is not 1:1
  • What is a signal transducer?
    • receptors?
    • ligands?
    • downstream kinases?
  • RECEPTORS
    • Should ligand-gated ion channels be given receptor parents?
    • Should receptors be divided into 'transport receptors' that deliver nutrients into the cell, and 'signaling receptors' that transduce a signal.
    • Should the RECEPTOR terms be kept in FUNCTION, or PROCESS terms created to replace them?
      • signal reception (synonym: receptor)
      • primary signal transduction (synonym: receptor)
  • receptor-mediated endocytosis
    • How should receptor-mediated endocytosis be linked to signaling?
    • Is the first step in RME (receptor binding to nutrient/ligand) signal transduction?
    • Where do RME receptors (eg LDLR, transferrin receptor) fit in?
  • CELL CYCLE CHECKPOINTS (Val)
    • How do the checkpoints fit with the signal transduction terms.
    • Val has been in contact with Paul Russell about checkpoints: He said: the term DNA damage checkpoint would represent the gene products involved in signalling only.
  • Connecting the signaling terms and response terms

SourceForge Signaling Items

receptors and HAS_PART: http://sourceforge.net/tracker/?func=detail&aid=3103037&group_id=36855&atid=440764

receptor signaling proteins: http://sourceforge.net/tracker/?func=detail&aid=3101487&group_id=36855&atid=440764

purigenic receptors: http://sourceforge.net/tracker/?func=detail&aid=3088079&group_id=36855&atid=440764

MAPKKK cascade: http://sourceforge.net/tracker/?func=detail&aid=3085150&group_id=36855&atid=440764

Calcium-mediated signaling; http://sourceforge.net/tracker/?func=detail&aid=3077377&group_id=36855&atid=440764

morphogen: http://sourceforge.net/tracker/?func=detail&aid=3026415&group_id=36855&atid=440764

norrin signaling (example of non 1:1 receptor:ligand signaling) : http://sourceforge.net/tracker/?func=detail&aid=3041964&group_id=36855&atid=440764

hormone and hormone receptor signaling: http://sourceforge.net/tracker/?func=detail&aid=2974691&group_id=36855&atid=440764


Connecting up receptors and intracellular modules:

wnt signaling via hippo cascade: http://sourceforge.net/tracker/?func=detail&aid=3032564&group_id=36855&atid=440764

wnt signaling via JNK cascade: http://sourceforge.net/tracker/?func=detail&aid=3023417&group_id=36855&atid=440764

PI3K/AKT signaling: http://sourceforge.net/tracker/?func=detail&aid=3008502&group_id=36855&atid=440764


Minutes

(preliminary version)

SO-Sandra Orchard; VW-Val Wood; AT-Andrea Townsend-Nicholson; RL-Ruth Lovering; RF- Rebecca Foulger; PD-Peter D’Eustachio; MH-Midori Harris; PR-Paola Roncaglia; VK-Varsha Khodiyar; ST-Susan Tweedie; AC-Andrew Chat-ra, SJ-Steve Jupe


Wed morning session

Where does 'signaling' begin and end, and what should 'signaling' include?

Questions

  • Q: Does signaling include synthesis (transcription, translation, processing) of the ligand and receptor?
  • Q: Does signaling include transport of the ligand and receptor to the correct place?
  • Q: Where does ligand processing take place?

Discussions

  • Because 'signaling' is so broad, and includes multi-organism signaling, nerve transmission etc, it's hard to find a common start point.
  • We need to distinguish 'signaling' from 'signal transduction', which starts with a receptor receiving a signal (E.g. ligand-receptor binding).
  • SO: There's overlap between the end of one pathway and the start of the next if you start with 'ligand production'.
  • For signaling between cells, its important to know which cell the signal comes from (we already have terms such as mesoderm-endoderm cell signaling).
  • For some pathways, the signal is synthesised in response to a stimulus. In some pathways secretion is the regulatory point (eg insulin receptor signaling where insulin is made in advance and secreted upon glucose stimulus). For some pathways, the stimulus is the signal (eg nutrient sensing).

Decisions

  • Decision: signaling (at molecular level) begins with an active signal (we need to look how this fits in with multi-organism signaling and check this with specific examples).
  • Decision: definition of signal agreed and added to comment field in signaling term.
  • Decision: transport of signal between cells is included in signaling (this fits with terms such as endocrine signaling, paracrine signaling etc).
  • Decision: 'regulation of signaling' does not make much sense as a term. Regulation of 'signal transduction' may be better. What we mean is 'regulation of a signaling process'.
  • 'Decision: AT: GO needs a 'glossary' of terms for things like 'signal' so we can standardise each definition. This should go on the signaling wiki.

Actions

  • AI: Can 'generation of a signal involved in cell-cell signaling ; GO:0003001' be obsoleted/moved/broken into smaller processes? Need to look at children. It would be better if the term could be broken down into 'signal secretion, signal transport, signal processing' etc.




Endocrine vs Paracrine slide 8

Questions

  • Q: Should we distinguish between endocrine and paracrine signaling in GO and add these as new terms (paracrine signaling is currently a narrow synonym of cell-cell signaling)?


'Discussions

  • Endocrinologist reaction that Stan L contacted was mixed, some said fine to split, some said not possible to tell experimentally.
  • AT – can you really distinguish between auto and para? These types of signaling are shown in intact organisms, not in cell based assays.
  • Decision: yes is it worth adding these but we should also add a comment that they should be used with care.
  • AI: make new terms for endocrine signaling and paracrine signaling: BECKY.



signal transduction, signal transducer activity and signaling pathways

Questions

  • Q: Where does signal transduction start and end?

Discussions

  • RF: A signaling pathway ends with regulation of a downstream cellular process.
  • There was a lot of discussions about whether transcription factors (TFs) should be part of a signaling pathway. AT said they should. ST said 'foxo' should be returned in a search for members of the insulin receptor signaling pathway.
  • Do we need the term 'signal transducer'? The general consensus is that this term could be obsoleted. We need to look at children though, and how this would fit with the proposal to change 'receptor signaling protein activity' to 'signal transducing protein kinase activity'.

Decisions

  • It's ok for the ligand to be annotated to 'signal transduction'.
  • Signal transduction begins with reception of a signal (E.g. ligand-receptor binding) and ends with regulation of a downstream process.
  • TFs are part of a pathway (eg part of signal transduction) because they pass the signal onto downstream RNA polymerase and regulate transcription.
  • Glycogen Synthase (GS) is NOT part of the insulin receptor signaling pathway because it is performing the act of glycogen synthesis. It is not passing the signal on.
  • AI: Check with David and Karen that this idea fits with transcription overhaul.

Wed afternoon session

initiation of signal transduction and termination of signal transduction (see slides 17-20)

Questions

  • Q: Where does 'initiation of signal transduction' end?... with an activated receptor, or when an activated receptor has passed the signal to the next signaling molecule?
  • Q: What does 'termination of signal transduction' include? Attenuation or just negative regulation?

Discussions

  • The comment and definition are not consistent: The comments say that both ligands and receptors can be annotated to 'initiation of signal transduction', but for many (eg signal transduction by light), there is no ligand gene product to annotate.
  • VW:termination of signal transduction is like saying the last step in a pathway x which is meaningless.
  • Discussion of whether the signal is always extracellular (AT: says Oxford Dictionary def of signal transduction includes an extracellular ligand). VW: SIN doesn't start with an extracellular ligand.
  • Decision: The general conclusion was that these terms aren't useful and it makes more sense to remove them than tighten up the definitions. Only 6 annotations have been made with initiation of signal transduction and these are better captured using function terms e.g. 'signal induction by light' would be better captured with a term 'light activated receptor activity'. There are no annotations to 'termination of signal transduction'.
  • AI: obsolete initiation of signal transduction and termination of signal transduction.

impact on linking pathways to response to terms (see slide 24)

Questions'

  • Q: We want to link pathways to the appropriate 'response to' terms in GO e.g insulin receptor signaling pathway is_a cellular response to insulin stimulus. Now the the ligand is included in the signaling pathway is it a problem that insulin itself would be returned in a search for 'cellular response to insulin stimulus'?

Discussions: Members of the insulin receptor signaling pathway from receptor-downwards, should be returned in a search for 'response to insulin'. The only way to get around it would be to create child terms for every pathway, beginning from an active receptor. The only difference between this and the parent term would be the ligand. In practise this would be a huge task and confusing to annotators. It would also be problematic when connecting the pathways to other events: do you choose the whole pathway, or just from receptor downwards?

There wasn't complete agreement about this (AT: didn't want see insulin in this search and thought it could be misleading but SO: would be very surprised if it wasn't there) but it was agreed that having a few potentially unexpected ligands in these searches was a small price to pay for the value of retaining the link to the response to term. We also agreed that the language sound ugly (PD) but there wasn't much we could do about it.

Decisions

  • We can live with insulin being returned in a search for 'response to insulin stimulus'.

Use of cascades in GO (see slides 45-60)

GO currently has a series of cascade that represent intracellular modules of signaling pathways.

Questions:

  • Q: Is cascade an appropriate name for these sub-pathways?
  • Q: Does there have to be amplification for it to be a 'cascade'?

Discussions: SO and ST expressed concern that the word cascade implies an amplification of the signal and this in not seen in all processes currently called cascades. It was accepted that while this usage is not universal it may be better to use a different word to avoid potential confusion. A PubMed search by AD suggested that cascade was less popular than pathway in this context. Various options were suggested: sub-pathway, AC and PR: pathway, AT: cartridges, VK: intracellular module but keep cascade for usage where there is amplification.

AI: Canvas wider opinion for best name



Receptors that bind multiple ligands, and ligands that bind multiple receptors

Problem

  • To have the activity of an EGFR in function, the GP has to bind EGF and pass the signal on. In process, we are not consistent about whether we describe an EGFR as a receptor that binds EGF OR as the gene product that's called EGFR and can bind to a number of physiological ligands.
  • To create process-function links, we have to be consistent.

Questions

  • Does the end response depend on the ligand or receptor? (AT says both)

Decisions We came up with the proposed ontology structure:

       EGF-activated signaling pathway

--[isa]EGF-activated pathway by FGFR --[isa]EGF-activated pathway by EGFR

EGFR signaling pathway --[isa]EGF-activated pathway by EGFR --[isa]FGF-activated pathway by EGFR

FGFR signaling pathway --[isa]EGF-activated pathway by FGFR --[isa]FGF-activated pathway by FGFR

  • This allows you to go in via the receptor or the ligand, depending on the experiment.
  • In process 'EGFR' is the gene product called EGFR (this could get a PRO:ID).

Decision:

  • Rename 'epidermal growth factor receptor activity' to 'epidermal growth factor-binding receptor activity' (with a HAS_PART EGF binding relationship).

AI: Run this proposal past Chris Mungall.

Notes

Definition Adjustments:

  • GO terms
    • Signaling: The entirety of a process in which information is transmitted. This process begins with an active signal and ends when a response has been triggered.
    • Signal transduction: The cellular process in which a signal is conveyed to trigger a change in the activity or state of a cell. Signal transduction begins with reception of a signal, e.g. a ligand binding to a receptor or receptor activation by a stimulus such as light, and ends with regulation of a downstream process,e.g. regulation of transcription or regulation of a metabolic process. Signal transduction covers signaling from receptors located on the surface of the cell, and signaling via molecules located within the cell. For signaling between cells, signal transduction is restricted to events at and within the receiving cell.
  • Glossary of ancillary terms
    • Signal: any variable property or parameter that serves to convey information, and may be a physical entity such as a gene product or small molecule, a photon, or a change in state such as movement or voltage change.

Obsoleted Terms:

  • initiation of signal transduction
  • termination of signal transduction

Moved Terms:

New Terms: