Re: [go-friends] [go-discuss] Announcement: three new relationship types to be added to core GO file

[Alan Ruttenberg]

[narrowed cc]

With this definition, to the extend that I understand stages to be temporal parts of the histories of organisms, biological processes such as the ones mentioned should land up being inferred to be parts of them. That's because of the definition of history, one that was chosen so that we can move freely between continuants and their lives, since history-of is 1:1.

"A history is a process that is the sum of the totality of processes taking place in the spatiotemporal region occupied by a material entity or site, including processes on the surface of the entity or within the cavities to which it serves as host. (axiom label in BFO2 Reference: [138-001])"

By Chris' note it seems that happens during implies that the spatiotemporal regions are in a part-of relation.

In practice, that means that happens_during can be used, but there shouldn't be a surprise if the relata are infered to be part of.

Phil's example is better, since wearing pants has pants as participant and so the spatial region of the process isn't within the spatial region of the wearer. So part_of wouldn't hold. 

If the hierarchy is:

process

 history

  stage

 biological process

and make biological process disjoint from 

That will work fine, I think (but will think some more), as long as no stage amounts to being a biological process (which seems unlikely, IMO).

Your extract doesn't make clear why a biological process that happens during a stage wouldn't be part of it. Can you explain the reasoning?

-Alan

On Thu, May 9, 2013 at 5:10 PM, David Osumi-Sutherland <dj...@gen.cam.ac.uk> wrote:

On 9 May 2013, at 22:16, Alan Ruttenberg wrote:

Hi,

Thanks. 

There are three words you use here. Phases stages and processes. I can  look up biological process. Can you tell me where I can find the other two please?

From a paper I've been working on:

For any single species, many events during development occur in an invariant order.  Developmental biologists, particularly those working on model organisms, traditionally measure the progress of development in a single species relative to the occurrence of some standard series of easily and reliably score-able events whose order is invariant [1, 4, 5]. These events are commonly the beginning, end or some easily score-able key point during a major developmental process, but may also simply be an easily score-able conjunction of morphological features. These events define boundaries between developmental stages, relative to which the timing of gene expression, developmental processes and the birth, death or transformation of anatomical structures can be recorded.

Stage series are not limited to development of a whole animal.  They can be used for any developmental process with events that occur in an invariant order.  Drosophila oogenesis, for example, has a well defined stage series [6], and named, rather than numbered stages have been defined for wing development [7] and the development of ommatidia from clusters of precursors [8].

Here are some examples:

A standard stage series for Xenopus leavis development:

http://www.xenbase.org/anatomy/alldev.do

oogenesis stages

http://ars.els-cdn.com/content/image/1-s2.0-S1097276500800817-gr1.jpg

Cell cycle phases:

http://en.wikipedia.org/wiki/Cell_cycle#Phases

(I think it is reasonable to thing of cell cycle phases as stages.  The term phase is presumably preferred over stage for dividing a cycle.)

BTW - I think all of these are subclasses of BFO:process, as currently defined in BFO2.  But this may reflect a difference between the BFO2 conception of process and the usage of this term by biologists.

Cheers,

David

[Alan Ruttenberg]

If you don't think stages are temporal parts of histories (and will include something that justifies that in the definition) then I think the chain -> spatiotemporal region , part-of should be changes to something like projects onto temporal region o temporal part of.

-Alan

[David Osumi-Sutherland]

Hi Alan,

The way I've been thinking about this,  stages are not simply a way to divide up histories.   The term cell cycle clearly doesn't refer to a history, it refers to a cyclical process that all dividing eukaryotic cells go through.  Cell cycle phases are stages of this process.  Many things happen during the time that a cell is in a particular phase of its cycle that are not part of the cell cycle process. 

Many embryonic stages are defined with reference to developmental processes and it can be useful to relate the timings of these processes (relative to the history of the whole animal*) directly.  For example, one might want to record that head involution happens during dorsal closure - but there can be no part relationship as there is no spatial overlap between the structures involved.

One interesting question is whether part_of between processes entails happens_during (rather than the inverse, which we've been discussing).  I've been assuming (perhaps naively) that we can use Allen relations and their derivatives directly between processes, stages or histories.  If happens_during corresponds to Allen:during then this inference of happens_during from parthood between processes (or stages or histories) is wrong.  But if it corresponds to Allen:sfo (see https://docs.google.com/document/d/1FudWR9lWdU5pIEUG5XtTMjn91k-jSoYImPIKZA3a_xY/edit) I *think* it is safe

Cheers,

- David

[*One thing that's been missing in this discussion, is any mention of a reference process or history in which class level assertions about relative timings can make sense.]

--
You received this message because you are subscribed to the Google Groups "bfo-owl-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email to bfo-owl-deve...@googlegroups.com.
To post to this group, send email to bfo-ow...@googlegroups.com.
Visit this group at http://groups.google.com/group/bfo-owl-devel?hl=en-US.
For more options, visit https://groups.google.com/groups/opt_out.
 
 

[David Osumi-Sutherland]

Sharing the following figure in order to make discussion about Allen relations more transparent. Allen:sfo refers to "starts or finishes or during" - I contend that part_of between processes necessarily entails one of these three.

[Alan Ruttenberg]

On Fri, May 10, 2013 at 10:43 AM, David Osumi-Sutherland <dj...@gen.cam.ac.uk> wrote:

Hi Alan,

The way I've been thinking about this,  stages are not simply a way to divide up histories.   The term cell cycle clearly doesn't refer to a history

Not surprising. History is a technical term defined by BFO 2. 

 

, it refers to a cyclical process that all dividing eukaryotic cells go through.  Cell cycle phases are stages of this process.  Many things happen during the time that a cell is in a particular phase of its cycle that are not part of the cell cycle process.

That makes sense as well. But the way you talk about it above, it sounds like the processes are part of the phases. I'm not trying to challenge you on the way people talk or conceptualize this. I'm just trying to understand what it means in terms of the upper level ontology we are working on. 

Many embryonic stages are defined with reference to developmental processes and it can be useful to relate the timings of these processes (relative to the history of the whole animal*) directly.  For example, one might want to record that head involution happens during dorsal closure - but there can be no part relationship as there is no spatial overlap between the structures involved.

Can you give me some references of defs so I can read and understand  dorsal closure and involution? Are both these processes, stages, or one of each?

I think that happens_during as a relation is well-defined. What I'm really trying to understand is what stages are, what biological processes are (how you and others understand them) and then recapitulate the logic that says the biological processes are never part of, nor equivalent to, the stages.

 

One interesting question is whether part_of between processes entails happens_during (rather than the inverse, which we've been discussing).  

I think you are asking whether p1 part_of p2 => p2 happens_during p1

I believe so, but should check. Part-of implies located in. Located in between occurrents implies the corresponding (spatiotemporal) regions are part_of each other  (but not the other way around). spatiotemporal inclusion requires spatial inclusion at each time, and temporal inclusion.

 

I've been assuming (perhaps naively) that we can use Allen relations and their derivatives directly between processes, stages or histories.  

Yes to processes and histories (derivatives from their temporal regions). When I better understand stages I'll have an opinion about that.

 If happens_during corresponds to Allen:during then this inference of happens_during from parthood between processes (or stages or histories) is wrong.  But if it corresponds to Allen:sfo (see https://docs.google.com/document/d/1FudWR9lWdU5pIEUG5XtTMjn91k-jSoYImPIKZA3a_xY/edit) I *think* it is safe

Can't see that document. Permissions.

Best,

Alan