Ontology meeting 2013-11-07

From GO Wiki
Jump to navigation Jump to search

Attendees: Jane, Paola, Becky, David, Heiko, Harold, Tanya, David OS,

Minutes: Becky

CARRY OVER FUNDING

We MIGHT have funds for a break-out meeting before the end-of-Feb (if we get roll-over funding). We could use that to do an ontology modelling (LEGO/modular annotation/COMA) meeting, or other potential areas listed here. Since it's not long away, need to get a list of who is available when, and decide a meeting based on those people. Where should the meeting be?

A Protege plug-in hackathon has been requested by Helen P, so we could combine the two meetings in Cambridge? AI: Paola to set up doodle-poll to find suitable dates.


Catalytic activity node review at Texas GO meeting?

Spurred by 10/29/13 email to Tanya from Jim:

...
The general issue of substrate specificity and ontology terms seems to be coming up 
in my term requests and I'm wondering if it's something where more general rules 
should be discussed, if they don't already exist.

Also, related to that, there are some very good senior enzymologists here at TAMU 
who have done things like organize enzyme conferences. Maybe we could have some 
consultations during the Spring GO meeting if they are around.

Jim

In alot of cases, you can't distinguish between the 2- or 3-digit EC number (parent) terms and the 4-digit EC number (child) terms. It's a substrate specificity problem. Agreed that we should take advantage of the experts in Texas (mid-March). Before the end of the year, we should highlight the issues.

Potential questions:

  • Is there a better (non-EC?) way we should group GO enzyme terms?

We could also ask these enzyme experts about the assays using an artificial substrate. We agreed we would annotate to the NATURAL reaction, not the artificial one. We'll need comments in GO terms to direct the annotators.

AI: Tanya to reply to Jim (DONE), and create JIRA ticket to summarize the issues. (DONE) GO-213

Template for RESPONSE TO STRESS GO terms

See SF item from Ingrid

Chris says: We have a small ontology of environmental conditions (oxidative stress etc) we use for making logical definitions of these terms. Time to add to production?

Q: What needs to be done next, and where is the environmental condition ontology?

Suggest changing our current definition of stress terms from:

[Term]
id: GO:0034976
name: response to endoplasmic reticulum stress
namespace: biological_process
def: "Any process that results in a change in state or activity of a cell (in terms of movement, secretion, enzyme production,  gene expression, etc.) as a result of a stimulus indicating endoplasmic reticulum (ER) stress. ER stress usually results from the accumulation of unfolded or misfolded proteins in the ER lumen." [GOC:mah]
synonym: "ER stress response" EXACT []
synonym: "response to ER stress" EXACT []
is_a: GO:0033554 ! cellular response to stress

to:

... as a result of a stress stimulus at the endoplasmic reticulum (ER).
... as a result of a stress stimulus acting at the endoplasmic reticulum (ER).
... as a result of a stimulus disrupting the endoplasmic reticulum (ER).


environmental ontology lives in extension directory [1]. It was built semi-automatically a while ago, taking all terms with 'stimulus' in.

  • Subbranch: response to chemical (already have ChEBI-based logical definitions)
  • Subbranch: response to stress (stress, heat, cold etc).

This may be alot of work to make templatable, as we'd need to add to the environmental condition ontology. It's hard to come up with a formal definition of 'stress'.

As an aside, need to standardise terms since we have some 'response to x' and 'response to y stimulus'. Add or remove 'stimulus' from all these terms.


Follow-up: TG TEMPLATE FOR ORGANELLE PART (LUMEN, MEMBRANE)

Chris, David OS and Heiko to report from last week. (see http://wiki.geneontology.org/index.php/Ontology_meeting_2013-10-31#Follow-up:_TG_TEMPLATE_FOR_ORGANELLE_PART_.28LUMEN.2C_MEMBRANE.29)

discussed how to use 'surrounds' location. Ambiguous whether it's part_of or not. Could consider using 'bounds' instead. Need guidance notes.

Could add 'secreted_by' relation in annotation extension.

New OE

Quick run through of new imports and visualization. Any tips to share?


Stemming from question of adding 'luteinizing hormone' in GO

See email thread. Chris says: "GO eds - we will want to switch IDs in the production logical definitions before this, to check the inferences. That's another external ontology imports file - not a big deal. Discuss on thursday?"


GEFs and GAFs

GEF activity is currently a child of 'GTPase regulator activity':

GTPase regulator activity ; GO:0030695
--[isa]guanyl-nucleotide exchange factor activity ; GO:0005085
guanyl-nucleotide exchange factor activity ; GO:0005085 (GEF)
Stimulates the exchange of guanyl nucleotides associated with a GTPase. Under normal cellular physiological conditions, the concentration of GTP is higher than that of GDP, favoring the replacement of GDP by GTP in association with the GTPase.

Does GEF fit the bill to be a GTPase REGULATOR? GEF alters binding affinity, not the catalytic activity directly. So under normal physiological conditions, it positively regulates GTPase activity, but is it a REGULATOR?

stems from SF ticket from Val





Review of template requests in Jira

We'd like to group these JIRA issues, but you can only have 2 levels in JIRA (issues and sub-issues).



MENGO Requests

Extra issue from Jane

Need to be able to capture cases in microbes (E.g. E-coli) to annotate synthetic pathways, as requested by Trudie. Should this be in a separate ontology? Is any of it applicable to GO?

One request is for 'MF involved in X pathway' terms. Since there's a finite number of pathways, this should be okay.

What's less clear is how to annotate the modified pathways- they use a natural pathway and tweak a few steps to see how it affects the overall output. LEGO would allow them to construct their specific pathway variations. It's analagous to annotating phenotypes.