A signaling call to catch up on where the signaling revisions are up to and what needs to be done next.
MONDAY 28TH OCTOBER 2013
- 3:30pm UK time
- 11:30am East Coast
- 8:30am West Coast
Attendees: Becky, David H, David OS, Chris, Tanya, Judy
AGENDA: OUTSTANDING SIGNALING ISSUES
- 1 INTRACELLULAR CASSETTES
- 2 LINKING INTRACELLULAR CASSETTES TO UPSTREAM RECEPTORS
- 3 RECEPTOR AGONISTS
- 4 RECEPTOR BINDING
- 5 SIGNAL GENERATION
- 6 ACTIVATED VERSUS MEDIATED
- 7 CARBOHYDRATE-MEDIATED SIGNALING
- 8 LIGAND RECEPTOR SIGNALING PATHWAYS
- 9 TOO BROAD FOR MANUAL ANNOTATION
- 10 LIGAND-GATED ION CHANNELS
- 11 Minutes
Need to define the start and stop points of these, and clarify what we mean by them. []
LINKING INTRACELLULAR CASSETTES TO UPSTREAM RECEPTORS
Is it worth creating GO terms to link the signaling pathways with the intracellular cassette? What would we annotate to them?
Or do we just wait for LEGO?
The term 'receptor agonist activity ; GO:0048018' has only one child (death receptor agonist activity). How can we expand this node?
If we have 'EGFR agonist', do we mean a ligand that activates the GENE PRODUCT 'Egfr'? If so, it can't be connected to any receptor term (e.g. it can't have 'positively_regulates EGF-activated receptor activity, because that's not the same thing).
If we have 'EGFR agonist' which positively_regulates 'EGF-activated receptor activity, it can ONLY be used to annotate EGF. Which seems restrictive and not useful.
Or should we capture this information in column 16 and forget about the links within the ontology?
When we say 'epidermal growth factor receptor binding ; GO:0005154' do we mean:
- Binding to the gene product 'Egfr'.
- Binding to a receptor that is activated by EGF
You can't bind to an activity. But you can bind to a GP that has an activity.
- Q: Is signal generation THE START OF signaling or UPSTREAM of signaling?
- Need to decide, for placement of GO:0003001, which says that generation of the signal is involved in signaling. This contradicts the current definition of signaling which begins with an active signal.
- Is biosynthesis of the signal (e.g. translation/sphingolipid biosynthesis/light generation) really part of signaling?
generation of a signal involved in cell-cell signaling ; GO:0003001 This process begins with the initial formation of the signal and ends with the mature form and placement of the signal.
signaling ; GO:0023052 This process begins with an active signal and ends when a cellular response has been triggered.
ACTIVATED VERSUS MEDIATED
- Need to standardise term names to 'activated' OR 'mediated'. E.g.:
- collagen-activated signaling pathway ; GO:0038065
- cytokine-mediated signaling pathway ; GO:0019221
Q: What do we mean by 'carbohydrate mediated signaling ; GO:0009756'. Is carbohydrate a ligand for a receptor?
LIGAND RECEPTOR SIGNALING PATHWAYS
- It's confusing to have 'ligand receptor signaling pathways'
- gurken receptor signaling pathway
- wnt receptor signaling pathway
- We could remove 'receptor' from the names:
- gurken signaling pathway (or gurken-activated signaling pathway)
- Wnt signaling pathway
This may mess up the growth factor receptor signaling pathways. And GPCRs. Is it: EGF-signaling pathway or 'EGFR signaling pathway'?
TOO BROAD FOR MANUAL ANNOTATION
Should these be made too broad for manual annotation?
signal transduction ; GO:0007165 signal transducer activity ; GO:0004871
LIGAND-GATED ION CHANNELS
QUESTION: Should these have parentage in the 'receptor' node of GO?
Alot of discussion about the intracellular signaling cassettes, and where they should stop and start. Suggested by David H, that PI3K signaling begins WITH PI3K. But a lot of annotations have been using the term more broadly to mean anything that signals THROUGH PI3K (including ligands, receptors, upstream activators and downstream effectors).
AI: Becky to have a play with the ontology and broaden PI3K signaling to mean anything that signals through PI3K (will need to decide whether to move out from under intracellular signal transduction). Then create a separate module for PI3K signaling which will have a HAS_PART link to the whole pathway.
Chris presented a demo of how this links to LEGO. For more details, see: https://docs.google.com/document/d/1TV8Eb9sSvFY-weVZaIfzCxF1qbnmkUaiUhTm9Bs3gRE/edit#
Return to Signaling Main Page []