Ontology meeting 2012-11-08

From GO Wiki
Revision as of 08:37, 6 February 2013 by David (talk | contribs) (→‎FOLLOW-UP: ChEBI paper)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to navigation Jump to search


ATTENDEES: David, Heiko, Harold, Chris, Paola, Jane, Tanya


References up to date. Restructuring done by Tanya, passed back to David. Make sure your Dropbox folder has version 13 in it. David edited. There is version 14 now. Look at text. David is reworking figures.

Open Access - pay ok? David to check.

Tanya will invite other authors to Dropbox. Jane to send email addresses of missing. Deadline for response/comment: end of next week.

FOLLOW-UP: import/export template

David and Tanya to report. Added xps from Chris' file to go_xp_chebi.obo. Heiko to begin working on template when opportunity arises (I made a JIRA ticket for this, assigned to Heiko - Jane).

Jane to report briefly on non-entailed import/export links (see [1])

We've also just noticed that we're getting a lot of new non-entailed links in the report e.g.

EXISTS, NOT-ENTAILED	SubClassOf(GO:0032973 'amino acid export' GO:0006865 'amino acid transport')
EXISTS, NOT-ENTAILED	SubClassOf(GO:0033216 'ferric iron import' GO:0015682 'ferric iron transport')

Why? Need to add class axiom that import/export relations propagate over conjugate base. Chris and Heiko will do this, will probably use GO_REL ids as they are experimental.

There is another file in the experimental directory that needs reviewing. AI: Jane to edit.


Add to go_xp_chebi.obo.

FOLLOW-UP: instantiating inferred/removing redundant links

Was there a file of redundant links we needed to check? Also, what's the pipeline for this in the editors file - ad hoc?

Heiko will create file and Jane can review.

Maintenance: use assert implied functionality within Oort or can run separately. Heiko will send instructions.

CHEBI roles in crossproduct file

We currently have xps to CHEBI terms that are roles in the xp file. Do we keep them or not? YES and add more as necessary. David There are property chains in x_chemicals files to take care of inferences. If we keep them, should we allow TermGenie to create terms using such CHEBI terms? Yes, allow any ChEBI terms. Heiko For example, SF request [2].

  id: GO:0042493 ! response to drug
  intersection_of: GO:0050896 ! response to stimulus
  intersection_of: has_input CHEBI:23888 ! drug

Chris: If they're terms that are useful to people (and Helen in particular) and in scope for GO we should add them.

If we add a term, we should have a logical definition upfront, rather than doing this retrospectively and creating more work for ourselves.

We can always change the pattern later. E.g. we may later decide that instead of

'response to' some (entity and has_role chebi:'aromatase inhibitor')

we want

'response to' some (entity and capable_of some (negatively_regulates some 'GO:0070330 ! aromatase activity')

we can do that later

we can also do

(has_role chebi:'aromatase inhibitor') EquivalentTo capable_of some (negatively_regulates some 'GO:0070330 ! aromatase activity')

Related 'response to' question: Lots of requests for these types of terms where x = chemical. The chemical itself may not occur naturally. What do we think about the synthetic chemicals? Our annotation guideline: 'A gene product should be annotated with terms reflecting its normal activity and location.' Is something like 'response to GW 7647' something normal? Add these regardless of origin? It may be a pain to go back and pull out all of the synthetic chemicals (thinking of drugs now and that slew of terms requested by Diane). Thoughts?

  id: GO:1901593
  name: response to GW 7647
  namespace: biological_process
  def: "Any process that results in a change in state or activity of a cell or an organism (in terms of movement, 
  secretion, enzyme production, gene expression, etc.) as a result of a GW 7647 stimulus." [GOC:TermGenie]
  is_a: GO:0010033 {is_inferred="true"} ! response to organic substance
  intersection_of: GO:0050896 ! response to stimulus
  intersection_of: has_input CHEBI:70778
  relationship: has_input CHEBI:70778 {is_inferred="true"}

Jane: lots of proks respond to synthetic compounds, they rapidly evolve to deal with them so processes can be considered normal. Not so sure for more complex orgs, these are probably more phenotype-y.

Group: It's ok. We'll have it in GO in a controlled way. We can address issues in the future if we have the terms here.

GO chemical terms where chemical is_a mixture

Stemming from placement of 'phlobaphene metabolic process'. See email thread among some of us editors.

Phlobaphene = A mixture of flavonoid derivatives that forms the red pigment of plants.

  • I think we finally settled on using the single axiom: 'metabolism' SubClassOf has_participant some 'chemical entity'? Is this expressible in OBO though?

Solution: Place under 'metabolic process'. Paola Remove 'chemical substance M/B/C' terms in queue. Tanya

TermGenie template for 'cellular response to'

This type of term is frequently requested via SF. I think we've discussed its addition previously, but don't remember if we reached a resolution - can't find anything on wiki or email threads. Would it be possible to implement a sort of 'decision tree' for the non-ChEBI parents? Or shall we leave this for the non-template-able term requests procedure?

E.g. see this SF request: https://sourceforge.net/tracker/?func=detail&aid=3580560&group_id=36855&atid=440764 [Paola]

Chris will parse existing terms and create xp file for review. Need template: genus: cellular response to chemical

Endosomal transport

Follow up from last week, Tanya to report.

Endosomal transport: The directed movement of substances into, out of or mediated by an endosome, a membrane-bounded organelle that carries materials newly ingested by endocytosis. It passes many of the materials to lysosomes for degradation.

two is_a parents: vesicle-mediated transport and intracellular transport

Vesicle-mediated transport: A cellular transport process in which transported substances are moved in membrane-bounded vesicles; transported substances are enclosed in the vesicle lumen or located in the vesicle membrane. The process begins with a step that directs a substance to the forming vesicle, and includes vesicle budding and coating. Vesicles are then targeted to, and fuse with, an acceptor membrane.

An endosome is NOT is_a vesicle, some vesicles fuse w/ endosome (endocytic), many endosomes contain lumenal vesicles that are transported within the endosome to another organelle after the two organelles fuse.

All current children appear to involve vesicles, even if it doesn't explicitly say so in the definition. I can't find one that is specifically saying that there are substances moving into or out of an endosome without another organelle being involved.


  • leave as is
  • Tanya THIS ONE: create vesicle-mediated endosomal transport as of endosomal transport, put all current kids under this (parent can be removed from vesicle-mediated parentage), send email to dbs to either move or leave in place, still correct
  • Chris and Tanya will make logical def for endosomal transport

SF corner

disulphide bond enzymes

Is it okay to merge a process into a function? See [3]

Should be ok, won't break anything. People will 'lose' a process annotation. Jane will look for a process parent (protein folding) to make the function/process link for that function. Reexamine protein modification parentage.

Enzymes that can use more than one substrate

Are they separate molecular functions? See SF3571860 and GO:0016619.

GO:0016619 - David will handle. Probably obsolete this term and create two new ones, one for each substrate so that the two reactions are separated. Need a biochemist to consult. (Peter? Authors of the paper?)

SF3571860 - David will see what is available in Rhea and KEGG.

Change xps for amino acid GO terms?

So that they use CHEBI:33704 alpha-amino acid instead of ChEBI:33709 amino acid (see [4]). This corresponds better to what biologists think of as amino acids. If we do this, it might be best to just delete all current children of aa terms and re-infer.

Chris: is worried about this change David: prefers adding 'alpha amino acid metabolism' and putting appropriate kids under that. (others agree) TermGenie will FIX!

DONE Status for commit #5371: Success (TermGenie RULES! New relationships created as expected.)

   ID: GO:1901605 Label: alpha-amino acid metabolic process
   ID: GO:1901606 Label: alpha-amino acid catabolic process
   ID: GO:1901607 Label: alpha-amino acid biosynthetic process