XP:biological process xp cellular component: Difference between revisions

From GO Wiki
Jump to navigation Jump to search
No edit summary
 
(28 intermediate revisions by 2 users not shown)
Line 1: Line 1:
XPs defining biological processes in terms of cellular components; example:
XPs defining biological processes in terms of cellular components; example:


  [Term]
<pre>
  id: GO:0000054 ! ribosome export from nucleus
[Term]
  intersection_of: GO:0051168 {namespace=biological_process_xp_cellular_component} ! nuclear export
id: GO:0010020 ! chloroplast fission
  intersection_of: OBO_REL:results_in_transport_of GO:0005840 {namespace=biological_process_xp_cellular_component} ! ribosome
intersection_of: GO:0048285  ! organelle fission
intersection_of: OBO_REL:results_in_division_of GO:0009507 ! chloroplast


i.e. ribosome export from nucleus = nuclear export THAT results_in_transport_of ribosome
[Term]
id: GO:0043572 ! plastid fission
intersection_of: GO:0048285  ! organelle fission
intersection_of: OBO_REL:results_in_division_of GO:0009536 ! plastid
</pre>


(Experimenting with the use of namespace tags as an easy way to separate out xp content from main GO)
i.e.
 
* '''chloroplast fission''' <sub>=def</sub> '''organelle fission''' THAT [[results_in_division_of]] '''chloroplast'''
* '''plastid fission''' <sub>=def</sub> '''organelle fission''' THAT [[results_in_division_of]] '''plastid'''
 
This allows a ''reasoner'' to compute the [[is_a]] relation between the two BP terms (based on the definition and the is_a relation between the CCs)
 
== Availability ==


See the [http://www.geneontology.org/scratch/xps/ go/scratch/xps] directory
See the [http://www.geneontology.org/scratch/xps/ go/scratch/xps] directory
Line 20: Line 32:
* [http://www.berkeleybop.org/ontologies/obo-all/bp_xp_cellular_component/bp_xp_cellular_component.xp.html HTML]
* [http://www.berkeleybop.org/ontologies/obo-all/bp_xp_cellular_component/bp_xp_cellular_component.xp.html HTML]


Missing Link Report:
== Methods ==
 
An initial parse was made using [[Obol]]. The results were then manually vetted by cjm, and new xp defs manually added
 
=== mitochondrial X ===
 
Terms of the form "mitochondrial ''X''" are typically defined as
 
* An ''X'' that [[occurs_in]] a '''mitochondrion'''
 
Here ''unfolds_in'' is a relation between a process and the location of the participants throughout the process
 
E.g.
 
<pre>
[Term]
id: GO:0000963 ! mitochondrial RNA processing
intersection_of: GO:0006396 ! RNA processing
intersection_of: OBO_REL:occurs_in GO:0005739 ! mitochondrion
</pre>
 
See [https://sourceforge.net/tracker/?func=detail&atid=440764&aid=2020664&group_id=36855 this sf item] for clarification on RNA processing terms
 
=== Cellular X ===
 
Cellular X means X is executed at the level of the cell.
 
See [https://sourceforge.net/tracker/index.php?func=detail&aid=2104158&group_id=36855&atid=440764 sf item] for clarification
 
For now we define cellular X using the has_level relation, but this is difficult to define
 
=== biogenesis, organization etc ===
 
Awaiting clarification. See: [[Cellular_component_processes]]
 
Existing xp defs may be changed once this is resolved
 
The current plan is to get replace "organization and biogenesis" with "organization". See above page for details.
 
=== export terms ===
 
Examples:
 
<pre>
[Term]
id: GO:0000056 ! ribosomal small subunit export from nucleus
intersection_of: GO:0051168 ! nuclear export
intersection_of: OBO_REL:results_in_transport_of GO:0015935 ! small ribosomal subunit
 
[Term]
id: GO:0006608 ! snRNP protein import into nucleus
intersection_of: GO:0006606 ! protein import into nucleus
intersection_of: OBO_REL:results_in_transport_of GO:0030532 ! small nuclear ribonucleoprotein complex
 
[Term]
id: GO:0006626 ! protein targeting to mitochondrion
intersection_of: GO:0006605 ! protein targeting
intersection_of: OBO_REL:results_in_transport_to GO:0005739 ! mitochondrion
</pre>
 
== Examples ==
 
{|class="wikitable" style="text-align:center" border="1"
|-
!
! GO:0016043-cellular component organization
! GO:0022411-cellular component disassembly
! GO:0022607-cellular component assembly
! GO:0031555-transcriptional attenuation
! GO:0051168-nuclear export
!
|-
! GO:0005840-ribosome
| GO:0042254-ribosome biogenesis<sub>[[results_in_organization_of]]</sub>
 
| GO:0032790-ribosome disassembly<sub>[[OBO_REL:results_in_breakdown_of]]</sub>
 
| GO:0042255-ribosome assembly<sub>[[OBO_REL:results_in_formation_of]]</sub>
 
| GO:0031556-transcriptional attenuation by ribosome<sub>[[OBO_REL:mediated_by]]</sub>
 
| GO:0000054-ribosome export from nucleus<sub>[[OBO_REL:results_in_transport_of]]</sub>
 
|
! GO:0005840-ribosome
|-
! GO:0015934-large ribosomal subunit
| GO:0042273-ribosomal large subunit biogenesis<sub>[[results_in_organization_of]]</sub>
 
|
|
|
| GO:0000055-ribosomal large subunit export from nucleus<sub>[[OBO_REL:results_in_transport_of]]</sub>
 
|
! GO:0015934-large ribosomal subunit
|-
! GO:0015935-small ribosomal subunit
| GO:0042274-ribosomal small subunit biogenesis<sub>[[results_in_organization_of]]</sub>
 
|
|
|
| GO:0000056-ribosomal small subunit export from nucleus<sub>[[OBO_REL:results_in_transport_of]]</sub>
 
|
! GO:0015935-small ribosomal subunit
|-
! GO:0033279-ribosomal subunit
|
|
| GO:0042257-ribosomal subunit assembly<sub>[[OBO_REL:results_in_formation_of]]</sub>
 
|
|
|
! GO:0033279-ribosomal subunit
|-
!
!
|-
!
! GO:0016043-cellular component organization
! GO:0022411-cellular component disassembly
! GO:0022607-cellular component assembly
! GO:0031555-transcriptional attenuation
! GO:0051168-nuclear export
|}
 
== Results ==
 
=== Missing Parentage Reports ===
 
* [http://sourceforge.net/tracker/index.php?func=detail&aid=1939817&group_id=36855&atid=440764 missing links] 14 added. closed.
* http://sourceforge.net/tracker/index.php?func=detail&aid=2015221&group_id=36855&atid=440764 -- made after it was clarified CCs are anatomical structures. See also [[Cellular_component_processes]]. Open
* [http://sourceforge.net/tracker/index.php?func=detail&aid=2016121&group_id=36855&atid=440764 various] closed
* [https://sourceforge.net/tracker/?func=detail&aid=2713459&group_id=36855&atid=440764 2009/03]
 
== Curator Notes ==
===2008-11-13===
Committed some updates today:
*updated alt_ids to ids:
**18994 -> 43186
**32842 -> 42575
*updated GO term names
*started adding comments on individual xps
====Matters and questions arising====
I'm checking to make sure it's OK to change results_in_organization_of to results_in_structural_organization_of; the latter is in ro_test.obo. It does seem to be OK in general in that the definition of the parent 'cellular component organization' is consistent with the relation def, but I'll also look at each xp that uses this relation.
 
Update (Nov. 20): I've looked at the typedef stanza for results_in_structural_organization_of, and I don't think it can be used for cellular component organization as currently defined. The problem is that the 'rule' tag implies that results_in_structural_organization_of is relevant in a developmental context, and several GO editors have just convinced each other that not all cellular component organization is necessarily developmental. See the mailing list thread starting [http://fafner.stanford.edu/pipermail/ontology-editors/2008-November/000163.html here] for more gory details.
 
'''Q:''' The xps also use has_level (see 'cellular x' above), which isn't in ro.obo or ro_test.obo, and I'm not as sure what to do about it.
 
'''Q:''' I've put in a [https://sourceforge.net/tracker/index.php?func=detail&aid=2309391&group_id=36855&atid=418257 bug report] about how OBO-Edit 1 and 2 handle unrecognized relations (differently annoyingly) -- OBO-Edit 1.101 failed to load the file, reporting "unrecognized type results_in_organization_of", whereas OBO-Edit 2beta50 loads everything and automagically converts the results_in_organization_of relation to is_a in the cross-product panel. This worries me ... is it really the desired behavior?
 
update (18 Nov; relevant to 2 Qs above): emailed Chris to ask about relations without typedef stanzas in general, because it affects various things in OE (including the Graphviz viewer, as in this [https://sourceforge.net/tracker/index.php?func=detail&aid=2277324&group_id=36855&atid=418257 bug report])
 
'''Q:''' Is it correct to use unfolds_in for terms such as mitochondrial ornithine transport, when the transport def is "into, out of or within" the CC?
 
update (18 Nov): emailed Chris thusly:
<pre>
The bp-xp-cc file has several entries for '[some organelle] transport' terms, e.g.
 
[Term]
id: GO:0000066 ! mitochondrial ornithine transport
intersection_of: GO:0015822 ! ornithine transport
intersection_of: OBO_REL:unfolds_in GO:0005739 ! mitochondrion
 
The def of GO:0000066, however, is "The directed movement of ornithine, 2,5-diaminopentanoic
acid, into, out of or within a mitochondrion."
 
It seems incorrect to use unfolds_in when the transport def is "into, out of or within" the CC
-- the unfolds_in def says "the execution of P is spatially contained by C", and that doesn't
seem to cover the "into" or "out of" bits. I guess the actual differentiae would be (unfolds_in
GO:0005739 OR results_in_transport_to GO:0005739 OR results_in_transport_from
GO:0005739) ... but is that possible/allowable/desirable?
</pre>
 
Note: unfolds_in changed to occurs_in
 
Less pressing matters:
 
RO and ro_test.obo aren't consistent w.r.t. derives_from, which makes OBO-Edit throw a warning. I've ignored it for now, since derives_from is in RO, but would like to know which to use when these xps go live.
 
Even smaller thing: the inverse_of tags on the obsolete relations has_improper_part and improper_part_of cause an OE warning. Ignored.
 
(Midori)
 
===2008-11-18===
Might want to add a term for 'plasma membrane transport' (maybe 'transport across plasma membrane'as name or synonym) -- at present plasma membrane pyruvate transport GO:0006849, plasma membrane acetate transport GO:0006847, etc. don't have a common parent.
 
Uncovered a general problem with transport terms -- the xp-generating rules seem to make two kinds of mistakes with transport process terms that mention some organelle.
 
*One is the problem noted above, where [organelle] [substrate X] transport is taken to mean transport of X that occurs within the organelle
*Another is [organelle X] transport, taken to mean transport OF the organelle instead of transport into, out of, or within the organelle.
Maybe this is something that can be addressed by structured synonyms?
 
MAH to do:
*check definition and come up with a better name for GO:0016182 (endosome to synaptic vesicle budding)- DONE
*check GO:0006998 - name is nuclear membrane organization, but the term dates back to before we properly distinguished the nuclear envelope from nuclear membrane(s) - DONE
*similar to above, for mitochondrial membrane org, GO:0007006 - DONE
*check defs, and make names consistent between BP and CC for focal adhesion/hemi-adherens junction (GO:0005925 & GO:0007045) - as of Nov. 20, cell junction BP & CC terms need an overhaul (last done in 2001!) [https://sourceforge.net/tracker/?func=detail&atid=440764&aid=2320317&group_id=36855 SF item] opened.
 
(midori)
 
===2008-11-21===
Transport terms referring to membranes have a problem related to the organellar transport problem: [membrane] [substrate X] transport means transport of X across the membrane. I'm not sure unfolds_in really captures that accurately. In the xp file I've noted that maybe results_in_transport_through might be better, but is undefined.
 
MAH to do:
*Ask Jen & Tanya to look at existing BP & CC graphs for this one:
<pre>
[Term]
id: GO:0010215 ! cellulose microfibril organization
intersection_of: GO:0016043 ! cellular component organization
intersection_of: results_in_organization_of GO:0009549 ! cellulose microfibril
</pre>
*bring up cell projection biogenesis terms next time we meet for CC org
*For GO:0016183, check where synaptic vesicles coat, and whether that's considered part of the synapse
**also minor def text edit
*try to define GO:0016240 autophagic vacuole docking
 
(midori)
 
===2008-12-15===
Finished first pass through xp file! Next thing to do is collect the 'ok' ones in a file and mark it as ready to go. Then sort out remaining ones with more complicated comments ...
 
I've noticed two xps (below) where the term has no path to anatomical structure formation in the current graph, but the xp suggests anat. str. formation as a genus. Do we want to add parentage, or just leave it to the xp statements?
 
<pre>
[Term]
id: GO:0001325 ! formation of extrachromosomal circular DNA
intersection_of: GO:0048646 ! anatomical structure formation
intersection_of: OBO_REL:results_in_formation_of GO:0005727 ! extrachromosomal circular DNA
 
[Term]
id: GO:0001771 ! formation of immunological synapse
intersection_of: GO:0048646 ! anatomical structure formation
intersection_of: OBO_REL:results_in_formation_of GO:0001772 ! immunological synapse
</pre>
 
MAH to do:
* make BP & CC names and synonyms consistent for cytochrome b(c)1/resp chain complex III - DONE
* double-check ribosome biogenesis vs. macromolecular complex organization - checked, but there's still some work to do for CC organization etc. that's relevant to these xps
* check whether GO:0043165 should become cell outer membrane assembly, with CC assembly parentage - DONE
** same for GO:0009833 primary cell wall and GO:0009834 secondary cell wall - decided to leave these (and parent) as 'biogenesis' because the defs mention 'reactions and pathways'
* sort out heterochromatin formation/chromatin silencing -- some child term not consistent with parent changes
* biogenesis vs organization for GO:0032581 - DONE
 
(midori)
 
===2009-06-16===
Emailed to ontdev:
 
So I finally overcame my fear of OBO-Edit making mincemeat of unrecognized relations, and  loaded the bp-xp-cc imports into OE for the first time in months. The very first implied link in the list (upon Edit -> Assert implied links) raises a question:
 
  Assert: protein complex assembly is_a cellular macromolecular complex assembly
 
I found the explanation a bit bewildering, but I think I see where the implication is coming from. The OE Graphviz viewer image might help:
 
[[Image:Protein_complex_assembly.png]]
 
(Note - image not included in email)
 
We've deliberately created separate terms for protein complex assembly and cellular protein complex assembly (with parallel structures for macromolecular complex assembly and m. c. subunit organization) because some complexes are assembled at the multicellular-organism level. The circulating lipoprotein complexes -- HDL, LDL, etc. -- are examples.
 
So it seems that, instead of asserting the implied link, we actually have to move cellular component organization and cellular component biogenesis directly under biological process, not under cellular process. This will fix a bunch of other implied links too, I think.
 
(midori)
 
====follow-up June 16 ====
Chris:<br>
So the inferences here follow from definitions that use the has_level relation. For example, cellular process = biological_process that has_level cell.
 
I'm really not sure the best way of defining has_level. But it seems that if we use this relation in logical definitions for terms that say "cellular <BP>" then the reasoner can at least make us consistent, as it has in the case above.
 
Midori:<br>
What rang the alarm bells is that the reasoner was suggesting links to a "cellular" parent for terms that are not themselves "cellular <bp>", like the example I included:
 
Assert: protein complex assembly is_a cellular macromolecular complex assembly
 
Obviously we ought to fix the inconsistency that's spawning those finds, rather than add the links as suggested. The biological_process_xp_cellular_component.obo file doesn't actually have any xps for "cellular component organization" or "cellular component biogenesis"; I think the problem is that both are is_a "cellular process".
 
But if we just move "cellular component organization" and "cellular component biogenesis" up to be direct is_a children of "biological process", some of the descendants will lose paths to "cellular process". The simplest solution is just to add the "cellular process" parent back to terms such as "cellular macromolecular complex subunit organization". A lot of fairly specific terms would end up directly under "cellular process" this way (e.g. "cytoplasm organization", "synaptogenesis"). Maybe we can live with that, although it would be neater to have "cellular <BP>" terms to group them. The catch with the better-organized approach is names for some of the terms -- "cellular cellular component assembly" sounds absolutely ridiculous, and words can't express how confused our users would be.
 
Any thoughts?
 
Jane:<br>
I was actually going to suggest we made this change (i.e. move cellular component organization and cellular component biogenesis directly under biological process) too recently, because I wanted to add a MOP term somewhere underneath.
 
Can't remember what the term was, or what became of it though! I'll see if I can find it.


* http://sourceforge.net/tracker/index.php?func=detail&aid=1939817&group_id=36855&atid=440764
* http://sourceforge.net/tracker/index.php?func=detail&aid=2015221&group_id=36855&atid=440764




[[Category:Cellular component]]
[[Category:Cross Products]]
[[Category:Cross Products]]
[[Category:Internal Cross Products]]
[[Category:Internal Cross Products]]

Latest revision as of 18:50, 11 March 2011

XPs defining biological processes in terms of cellular components; example:

[Term]
id: GO:0010020 ! chloroplast fission
intersection_of: GO:0048285  ! organelle fission
intersection_of: OBO_REL:results_in_division_of GO:0009507 ! chloroplast

[Term]
id: GO:0043572 ! plastid fission
intersection_of: GO:0048285  ! organelle fission
intersection_of: OBO_REL:results_in_division_of GO:0009536 ! plastid

i.e.

This allows a reasoner to compute the is_a relation between the two BP terms (based on the definition and the is_a relation between the CCs)

Availability

See the go/scratch/xps directory

This xp file has been registered in the OBO xps metadata so it is available here:

Methods

An initial parse was made using Obol. The results were then manually vetted by cjm, and new xp defs manually added

mitochondrial X

Terms of the form "mitochondrial X" are typically defined as

* An X that occurs_in a mitochondrion

Here unfolds_in is a relation between a process and the location of the participants throughout the process

E.g.

[Term]
id: GO:0000963 ! mitochondrial RNA processing
intersection_of: GO:0006396 ! RNA processing
intersection_of: OBO_REL:occurs_in GO:0005739 ! mitochondrion

See this sf item for clarification on RNA processing terms

Cellular X

Cellular X means X is executed at the level of the cell.

See sf item for clarification

For now we define cellular X using the has_level relation, but this is difficult to define

biogenesis, organization etc

Awaiting clarification. See: Cellular_component_processes

Existing xp defs may be changed once this is resolved

The current plan is to get replace "organization and biogenesis" with "organization". See above page for details.

export terms

Examples:

[Term]
id: GO:0000056 ! ribosomal small subunit export from nucleus
intersection_of: GO:0051168 ! nuclear export
intersection_of: OBO_REL:results_in_transport_of GO:0015935 ! small ribosomal subunit

[Term]
id: GO:0006608 ! snRNP protein import into nucleus
intersection_of: GO:0006606 ! protein import into nucleus
intersection_of: OBO_REL:results_in_transport_of GO:0030532 ! small nuclear ribonucleoprotein complex

[Term]
id: GO:0006626 ! protein targeting to mitochondrion
intersection_of: GO:0006605 ! protein targeting
intersection_of: OBO_REL:results_in_transport_to GO:0005739 ! mitochondrion

Examples

GO:0016043-cellular component organization GO:0022411-cellular component disassembly GO:0022607-cellular component assembly GO:0031555-transcriptional attenuation GO:0051168-nuclear export
GO:0005840-ribosome GO:0042254-ribosome biogenesisresults_in_organization_of GO:0032790-ribosome disassemblyOBO_REL:results_in_breakdown_of GO:0042255-ribosome assemblyOBO_REL:results_in_formation_of GO:0031556-transcriptional attenuation by ribosomeOBO_REL:mediated_by GO:0000054-ribosome export from nucleusOBO_REL:results_in_transport_of GO:0005840-ribosome
GO:0015934-large ribosomal subunit GO:0042273-ribosomal large subunit biogenesisresults_in_organization_of GO:0000055-ribosomal large subunit export from nucleusOBO_REL:results_in_transport_of GO:0015934-large ribosomal subunit
GO:0015935-small ribosomal subunit GO:0042274-ribosomal small subunit biogenesisresults_in_organization_of GO:0000056-ribosomal small subunit export from nucleusOBO_REL:results_in_transport_of GO:0015935-small ribosomal subunit
GO:0033279-ribosomal subunit GO:0042257-ribosomal subunit assemblyOBO_REL:results_in_formation_of GO:0033279-ribosomal subunit
GO:0016043-cellular component organization GO:0022411-cellular component disassembly GO:0022607-cellular component assembly GO:0031555-transcriptional attenuation GO:0051168-nuclear export

Results

Missing Parentage Reports

Curator Notes

2008-11-13

Committed some updates today:

  • updated alt_ids to ids:
    • 18994 -> 43186
    • 32842 -> 42575
  • updated GO term names
  • started adding comments on individual xps

Matters and questions arising

I'm checking to make sure it's OK to change results_in_organization_of to results_in_structural_organization_of; the latter is in ro_test.obo. It does seem to be OK in general in that the definition of the parent 'cellular component organization' is consistent with the relation def, but I'll also look at each xp that uses this relation.

Update (Nov. 20): I've looked at the typedef stanza for results_in_structural_organization_of, and I don't think it can be used for cellular component organization as currently defined. The problem is that the 'rule' tag implies that results_in_structural_organization_of is relevant in a developmental context, and several GO editors have just convinced each other that not all cellular component organization is necessarily developmental. See the mailing list thread starting here for more gory details.

Q: The xps also use has_level (see 'cellular x' above), which isn't in ro.obo or ro_test.obo, and I'm not as sure what to do about it.

Q: I've put in a bug report about how OBO-Edit 1 and 2 handle unrecognized relations (differently annoyingly) -- OBO-Edit 1.101 failed to load the file, reporting "unrecognized type results_in_organization_of", whereas OBO-Edit 2beta50 loads everything and automagically converts the results_in_organization_of relation to is_a in the cross-product panel. This worries me ... is it really the desired behavior?

update (18 Nov; relevant to 2 Qs above): emailed Chris to ask about relations without typedef stanzas in general, because it affects various things in OE (including the Graphviz viewer, as in this bug report)

Q: Is it correct to use unfolds_in for terms such as mitochondrial ornithine transport, when the transport def is "into, out of or within" the CC?

update (18 Nov): emailed Chris thusly:

The bp-xp-cc file has several entries for '[some organelle] transport' terms, e.g.

[Term]
id: GO:0000066 ! mitochondrial ornithine transport
intersection_of: GO:0015822 ! ornithine transport
intersection_of: OBO_REL:unfolds_in GO:0005739 ! mitochondrion

The def of GO:0000066, however, is "The directed movement of ornithine, 2,5-diaminopentanoic
acid, into, out of or within a mitochondrion."

It seems incorrect to use unfolds_in when the transport def is "into, out of or within" the CC
-- the unfolds_in def says "the execution of P is spatially contained by C", and that doesn't
seem to cover the "into" or "out of" bits. I guess the actual differentiae would be (unfolds_in
GO:0005739 OR results_in_transport_to GO:0005739 OR results_in_transport_from 
GO:0005739) ... but is that possible/allowable/desirable?

Note: unfolds_in changed to occurs_in

Less pressing matters:

RO and ro_test.obo aren't consistent w.r.t. derives_from, which makes OBO-Edit throw a warning. I've ignored it for now, since derives_from is in RO, but would like to know which to use when these xps go live.

Even smaller thing: the inverse_of tags on the obsolete relations has_improper_part and improper_part_of cause an OE warning. Ignored.

(Midori)

2008-11-18

Might want to add a term for 'plasma membrane transport' (maybe 'transport across plasma membrane'as name or synonym) -- at present plasma membrane pyruvate transport GO:0006849, plasma membrane acetate transport GO:0006847, etc. don't have a common parent.

Uncovered a general problem with transport terms -- the xp-generating rules seem to make two kinds of mistakes with transport process terms that mention some organelle.

  • One is the problem noted above, where [organelle] [substrate X] transport is taken to mean transport of X that occurs within the organelle
  • Another is [organelle X] transport, taken to mean transport OF the organelle instead of transport into, out of, or within the organelle.

Maybe this is something that can be addressed by structured synonyms?

MAH to do:

  • check definition and come up with a better name for GO:0016182 (endosome to synaptic vesicle budding)- DONE
  • check GO:0006998 - name is nuclear membrane organization, but the term dates back to before we properly distinguished the nuclear envelope from nuclear membrane(s) - DONE
  • similar to above, for mitochondrial membrane org, GO:0007006 - DONE
  • check defs, and make names consistent between BP and CC for focal adhesion/hemi-adherens junction (GO:0005925 & GO:0007045) - as of Nov. 20, cell junction BP & CC terms need an overhaul (last done in 2001!) SF item opened.

(midori)

2008-11-21

Transport terms referring to membranes have a problem related to the organellar transport problem: [membrane] [substrate X] transport means transport of X across the membrane. I'm not sure unfolds_in really captures that accurately. In the xp file I've noted that maybe results_in_transport_through might be better, but is undefined.

MAH to do:

  • Ask Jen & Tanya to look at existing BP & CC graphs for this one:
[Term]
id: GO:0010215 ! cellulose microfibril organization
intersection_of: GO:0016043 ! cellular component organization
intersection_of: results_in_organization_of GO:0009549 ! cellulose microfibril
  • bring up cell projection biogenesis terms next time we meet for CC org
  • For GO:0016183, check where synaptic vesicles coat, and whether that's considered part of the synapse
    • also minor def text edit
  • try to define GO:0016240 autophagic vacuole docking

(midori)

2008-12-15

Finished first pass through xp file! Next thing to do is collect the 'ok' ones in a file and mark it as ready to go. Then sort out remaining ones with more complicated comments ...

I've noticed two xps (below) where the term has no path to anatomical structure formation in the current graph, but the xp suggests anat. str. formation as a genus. Do we want to add parentage, or just leave it to the xp statements?

[Term]
id: GO:0001325 ! formation of extrachromosomal circular DNA
intersection_of: GO:0048646 ! anatomical structure formation
intersection_of: OBO_REL:results_in_formation_of GO:0005727 ! extrachromosomal circular DNA

[Term]
id: GO:0001771 ! formation of immunological synapse
intersection_of: GO:0048646 ! anatomical structure formation
intersection_of: OBO_REL:results_in_formation_of GO:0001772 ! immunological synapse

MAH to do:

  • make BP & CC names and synonyms consistent for cytochrome b(c)1/resp chain complex III - DONE
  • double-check ribosome biogenesis vs. macromolecular complex organization - checked, but there's still some work to do for CC organization etc. that's relevant to these xps
  • check whether GO:0043165 should become cell outer membrane assembly, with CC assembly parentage - DONE
    • same for GO:0009833 primary cell wall and GO:0009834 secondary cell wall - decided to leave these (and parent) as 'biogenesis' because the defs mention 'reactions and pathways'
  • sort out heterochromatin formation/chromatin silencing -- some child term not consistent with parent changes
  • biogenesis vs organization for GO:0032581 - DONE

(midori)

2009-06-16

Emailed to ontdev:

So I finally overcame my fear of OBO-Edit making mincemeat of unrecognized relations, and loaded the bp-xp-cc imports into OE for the first time in months. The very first implied link in the list (upon Edit -> Assert implied links) raises a question:

 Assert: protein complex assembly is_a cellular macromolecular complex assembly

I found the explanation a bit bewildering, but I think I see where the implication is coming from. The OE Graphviz viewer image might help:

(Note - image not included in email)

We've deliberately created separate terms for protein complex assembly and cellular protein complex assembly (with parallel structures for macromolecular complex assembly and m. c. subunit organization) because some complexes are assembled at the multicellular-organism level. The circulating lipoprotein complexes -- HDL, LDL, etc. -- are examples.

So it seems that, instead of asserting the implied link, we actually have to move cellular component organization and cellular component biogenesis directly under biological process, not under cellular process. This will fix a bunch of other implied links too, I think.

(midori)

follow-up June 16

Chris:
So the inferences here follow from definitions that use the has_level relation. For example, cellular process = biological_process that has_level cell.

I'm really not sure the best way of defining has_level. But it seems that if we use this relation in logical definitions for terms that say "cellular <BP>" then the reasoner can at least make us consistent, as it has in the case above.

Midori:
What rang the alarm bells is that the reasoner was suggesting links to a "cellular" parent for terms that are not themselves "cellular <bp>", like the example I included:

Assert: protein complex assembly is_a cellular macromolecular complex assembly

Obviously we ought to fix the inconsistency that's spawning those finds, rather than add the links as suggested. The biological_process_xp_cellular_component.obo file doesn't actually have any xps for "cellular component organization" or "cellular component biogenesis"; I think the problem is that both are is_a "cellular process".

But if we just move "cellular component organization" and "cellular component biogenesis" up to be direct is_a children of "biological process", some of the descendants will lose paths to "cellular process". The simplest solution is just to add the "cellular process" parent back to terms such as "cellular macromolecular complex subunit organization". A lot of fairly specific terms would end up directly under "cellular process" this way (e.g. "cytoplasm organization", "synaptogenesis"). Maybe we can live with that, although it would be neater to have "cellular <BP>" terms to group them. The catch with the better-organized approach is names for some of the terms -- "cellular cellular component assembly" sounds absolutely ridiculous, and words can't express how confused our users would be.

Any thoughts?

Jane:
I was actually going to suggest we made this change (i.e. move cellular component organization and cellular component biogenesis directly under biological process) too recently, because I wanted to add a MOP term somewhere underneath.

Can't remember what the term was, or what became of it though! I'll see if I can find it.