Re: Behavior Ontology
Janna Hastings
Glad to see the appearance of the behavior ontology, it addresses a
much needed gap and I will definitely plan to link the emotion
ontology (EMO) to it for all the characteristic behaviors that
typically accompany various emotional states.
I have some small concerns -- which I will also raise on your tracker.
Most of these depend on how you are going to define 'behavior' which
I see is not defined yet, but I find the following don't immediately
appear to me to be types of behavior?
-- 'intelligence' and its children. Not only doesn't it seem to me
to be a kind of behavior, but also not a kind of cognition (its
immediate is-a parent). Don't we remain intelligent even if we are
not thinking about anything? It seems rather a capability
(disposition) than a behavior (process).
-- Under 'emotional behavior' you have a few things that don't seem to
fit e.g. 'pleasure', 'frustration', 'distress' -- these are rather
feelings or emotions than they are behaviors, right? On the other
hand some of the emotional behaviors, such as 'laughing behavior',
'crying behavior' and 'aggression behavior', fit very well.
-- 'consciousness' also seems strange as a behavior, although I would
admit to being tempted to agree with you that sleeping is a behavior
(which you've labelled 'asleep'), I prefer its alternate appearance as
'sleeping behavior' than as a type of consciousness.
We should perhaps try to meet (I'll be at WoMBO and ICBO, see you
there?) to discuss the EMO--NBO bridging relations.
Best wishes!
Janna
On Mon, May 30, 2011 at 6:20 PM, Chris Mungall <cjmu...@lbl.gov> wrote:
>
> I suggest carrying this discussion on the obo-behavior list, and that George makes a formal announcement of NBO there.
>
> I agree re disorders:
> http://code.google.com/p/behavior-ontology/issues/detail?id=2
>
> Add any further issues here:
>
> http://code.google.com/p/behavior-ontology/issues/list
>
> I'll add the tracker link to the obo metadata page
>
> On May 30, 2011, at 8:14 AM, Barry Smith wrote:
>
>> George
>>
>> Janna Hastings brought to my attention your Neuro Behavior Ontology on Bioportal. This overlaps with work we have been doing in Buffalo and Geneva on Mental Disease Ontology
>> http://www.jbiomedsem.com/content/1/1/10
>> and on Emotion Ontology (to be presented at ICBO) (see draft attached) funded by the Swiss Centre for Affective Sciences.
>>
>> Since behavior (taken to include cognition) is clearly a gigantic target, we are hoping that you will agree to collaborate, and that we can settle on a strategy for this as soon as possible. The first and most important question is: should the top level be, as you propose,
>>
>> behavior
>> pathological behavior
>> physiological behavior
>>
>> Already I can see problems from this choice, e.g. if 'cognition' is classified under the latter, but 'mental disorder' is classified under the former.
>>
>> Another question concerns the consistency with OGMS,
>>
>> http://code.google.com/p/ogms
>>
>> which we are trying to get established as the OBO Foundry standard for clinically related work, and which provides a precise meaning for 'disorder'. The issue relates to the fact that, on the OGMS view (and we believe on other generally accepted clinical views) a disorder is not a subtype of behavior.
>> With greetings
>> Barry
>> <emonto_hastings_icbo2011.pdf>
>
>
Peter Midford
This looks like a solid start, and I am happy to start using it with my spider work. I will take some time in the next week to work through these and send you some suggestions.
Peter
Peter E. Midford
Phenoscape Ontology Curator
Peter....@gmail.com
George Gkoutos
Great Peter ! Thank you very much !
Looking forward to suggestions terms etc
George
Alan Ruttenberg
> George
> I think the best strategy is for you and Janna to make a first pass at
> updating the upper level structures of the ontology, and then we can
> work on the details further down when you are done.
> Dividing into 'physiological' and 'pathological' in order to do
> justice to canonicalness is not, I believe, a good strategy. I
> suggest, rather, that you just re-label the physiological branch as
> 'behavior' and leave the pathological stuff for later. The first job
> is then to provide a coherent definition of 'behavior' as a
> BFO:process. We have a proposal already in
> http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3017014/:
>
> • Behavior O) =def. a Process having Processes as parts in which an
> Organism participates as agent in response to external or internal
> stimuli and following some pattern which is dependent upon some
> combination of that Organism's internal state and external conditions.
> (Derived from the Gene Ontology)
I can't tell why this doesn't effectively reduce to: "A process that
is part of an organism's life". The issues:
1) Agentive participation: define? (for, e.g. a nematode, or a
bacterium). You might do better adopting FMA's "coordinated expression
of genes" strategy.
2) external or internal stimuli = anything
3) some combination of that Organism's internal state and external
conditions = anything
4) dependent on: not used in ontological technical sense - what sense then?
5) Quantification of "having Processes as parts..." is unclear - are
all process parts such? some? Are there parts (even fiat) of behavior
processes that have no organism as participant?
Against my definition (which isn't a serious attempt, of course)
Is two people dancing a behavior? Or only each person dancing? If the
former then the process has to be some sum of organism's life.
Same issue if walls can be participants in behaviors.
I'm sure there are more.
Critically yours,
Alan
>
> With greetings
> Barry
>
>
>
> On Tue, May 31, 2011 at 12:23 PM, George Gkoutos
> <g.gk...@gen.cam.ac.uk> wrote:
>>
>> On 31 May 2011, at 15:28, Barry Smith wrote:
>>
>>> One more item for the pot:
>>>
>>> I believe that 'Neuro Behavior Ontology' is not a good name for this
>>> resource. If it is to be a Behavior Ontology, then let us call it that
>>> (and grab the 'BO' acronym before the brucellosis people beat us to
>>> it)
>>>
>>
>> Happy to do this - it is still driven by my focus on mammalian behavior but now with the sebrafish people taking up on this maybe it should be behavior ontology and use BO. If people are happy I can do this.
>>
>> George
>>
>>> BS
> ------------------------------------------------------------------------------
> Simplify data backup and recovery for your virtual environment with vRanger.
> Installation's a snap, and flexible recovery options mean your data is safe,
> secure and there when you need it. Data protection magic?
> Nope - It's vRanger. Get your free trial download today.
> http://p.sf.net/sfu/quest-sfdev2dev
> _______________________________________________
> Obo-behavior mailing list
> Obo-be...@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/obo-behavior
>
Barry Smith
<alanrut...@gmail.com> wrote:
> On Tue, May 31, 2011 at 1:18 PM, Barry Smith <phis...@buffalo.edu> wrote:
>> George
>> I think the best strategy is for you and Janna to make a first pass at
>> updating the upper level structures of the ontology, and then we can
>> work on the details further down when you are done.
>> Dividing into 'physiological' and 'pathological' in order to do
>> justice to canonicalness is not, I believe, a good strategy. I
>> suggest, rather, that you just re-label the physiological branch as
>> 'behavior' and leave the pathological stuff for later. The first job
>> is then to provide a coherent definition of 'behavior' as a
>> BFO:process. We have a proposal already in
>> http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3017014/:
>>
>> • Behavior O) =def. a Process having Processes as parts in which an
>> Organism participates as agent in response to external or internal
>> stimuli and following some pattern which is dependent upon some
>> combination of that Organism's internal state and external conditions.
>> (Derived from the Gene Ontology)
>
> I can't tell why this doesn't effectively reduce to: "A process that
> is part of an organism's life". The issues:
> 1) Agentive participation: define? (for, e.g. a nematode, or a
> bacterium). You might do better adopting FMA's "coordinated expression
> of genes" strategy.
'agent' is so basic, it is difficult to see how one would define it;
by swooping all the way down to bacteria you are contravening the
principle of low hanging fruit
> 2) external or internal stimuli = anything
actually not; nearly everything is excluded, e.g. because it is on the
wrong end of the galaxy
but we do need a definition (or at least an elucidation) of 'stimulus'
> 3) some combination of that Organism's internal state and external
> conditions = anything
again, actually not
> 4) dependent on: not used in ontological technical sense - what sense then?
dependent precisely our technical sense
> 5) Quantification of "having Processes as parts..." is unclear - are
> all process parts such? some? Are there parts (even fiat) of behavior
> processes that have no organism as participant?
>
I, too, am not fully happy with this 'having processes as parts' bit
BS
Alan Ruttenberg
I don't buy that argument. But in any case, I would take a longish
list of positive and negative examples as a reasonable surrogate.
> by swooping all the way down to bacteria you are contravening the
> principle of low hanging fruit
I am against defining a definitive upper-level class when it isn't
applicable. Restrict the definition to specific fruit if that's where
it's applicable. So define "animal behavior" rather than behavior if
you think the low hanging fruit are animals.
>
>> 2) external or internal stimuli = anything
>
> actually not; nearly everything is excluded, e.g. because it is on the
> wrong end of the galaxy
> but we do need a definition (or at least an elucidation) of 'stimulus'
I read it as circular. Stimulus = something that can elicit behavior.
Behavior is process in response to stimulus. I'd love to have
something better.
>
>> 3) some combination of that Organism's internal state and external
>> conditions = anything
>
> again, actually not
Please defend this position. First "state" and "condition" are not
defined. I take them as synonymous. The phrase then reduces to "any
combination of state internal or external", which afaict includes any
state whatsoever. What is not included in the "state" of everything?
>
>> 4) dependent on: not used in ontological technical sense - what sense then?
>
> dependent precisely our technical sense
Please elucidate. We use the term dependent currently to mean a very
few types of relation - for the most part existential dependence,
modulo arguments about whether boundary dependence is existential
dependence.
Alan