Re: Definition of Networks and/or Pathways: Ontologies of Cellular Networks:
Barry Smith
>Hi:
>I want to thank you all for organizing the meeting and helping me
>with logistics there (apologies for belated thanks as I have been
>swamped since then). I am looking forward to output from the
>meeting. I promised myself to put the following as a contribution
>to the definition of N&P on the Google group, but figure I will send
>directly to you all (already shared with Alan). Once again thanks
>for excellent meeting. Peter
>
>Definition of Networks and Pathways in biology: A network or pathway
>is a reduced (or simplified) subset of a larger system or
>process. This may be a subset of either the underlying reality
>(molecules and things) or a model state (estimate of the underlying
>reality). Whether or not the definition refers to the underlying
>reality of the model state depends on context. For example, the
>chemicals in the Krebs cycle and their distribution in the cell are
>a network or pathway which is a subset of a cell or
>organism. Alternatively, the Krebs cycle may represented by figures
>in a text book, the images on a web site, or the numbers in a
>computer model; these are a model state of the Krebs cycle network pathway.
I think Peter's definition is very much on the right track. However,
I would prefer to avoid the term 'subset', because if a network or
pathway is a subset of a system or process, then systems and
processes must be sets, and since (on the normal understanding at
least) sets are abstract entities existing outside space and time, we
would thereby lose what is essentially biological (dynamic) in
systems and processes.
Better to say that a network or pathway is a reduced (or simplified)
subpart of a larger system or process. This may be a subpart ...
Not entirely non-ugly, but better,nonetheless.
BS
>
>Commentary:
> * The purpose of the reduction (or subsetting) is usually (i)
> networks or pathways are easier to understand and analyze than the
> whole system, and (ii) networks and pathways are usually chosen to
> be as self contained as possible. However a network or pathway
> that is instantiated in isolation from the system that it is
> usually found it will usually display problematic behavior; in this
> sense networks and pathways are an artificial, but often
> scientifically useful, construct.
> * This definition of networks and pathways is framed in the
> engineering discipline of filtering. In this discipline systems
> are have three components. First, the underlying reality is the
> molecules and things of material and they evolve in space and
> time. Second, observations are human's only access to the
> underlying reality. Since there are always errors in measurement
> it is understood that humans never have a precise measure of the
> underlying reality. Third, a model is a human construct which is
> designed to improve measurement and analysis of underlying
> reality. The output and numbers of a model is called the model
> state. Thus a model state is a (hopefully optimal) estimate of the
> underlying reality which uses a combination of observations and
> human knowledge as input to the model.
>
>
>Peter Lyster PhD
>Center for Bioinformatics and Computational Biology
>National Institute of General Medical Sciences
>45 Center Dr. #2as.55k
>Bethesda, MD 20892
>(301) 451-6446
><mailto:lys...@mail.nih.gov>lys...@mail.nih.gov
>-----Original Message-----
>From: Barry Smith [mailto:phis...@buffalo.edu]
>Sent: Friday, March 21, 2008 3:40 PM
>To: phis...@buffalo.edu
>Subject: Ontologies of Cellular Networks: Final Reminders
>
>Dear colleagues,
>
>I am looking forward to seeing you all in Newark on Thursday.
>Registration begins at 9am.
>
>The final version of the Agenda is here:
>http://www.bioontology.org/wiki/index.php/Networks
>together also with instructions on how to get to the Newark Airport
>Hilton, where our meeting will take place. (Look for Junior Ballroom
>when you get there.)
>
>Presenters have been instructed to keep their remarks to 10 minutes
>or less, in order to leave ample time for discussion. All slides will
>be posted to the meeting wiki, and presenters are encouraged to send
>their slides to me in advance (in longer versions if preferred).
>
>All participants should feel free to send me any other supplementary
>material which they think might be useful for posting to the wiki.
>
>With greetings
>Barry
>http://ontology.buffalo.edu/smith
>
Robert Arp Ph.D.
Best,
Rob
On Wed May 7 11:19 , Barry Smith <phis...@buffalo.edu> sent:
>>lys...@mail.nih.gov','','','')">lys...@mail.nih.gov>lys...@mail.nih.gov
>>-----Original Message-----
>>From: Barry Smith [phis...@buffalo.edu','','','')">phis...@buffalo.edu]
>>Sent: Friday, March 21, 2008 3:40 PM
>>To: phis...@buffalo.edu
>>Subject: Ontologies of Cellular Networks: Final Reminders
>>
>>Dear colleagues,
>>
>>I am looking forward to seeing you all in Newark on Thursday.
>>Registration begins at 9am.
>>
>>The final version of the Agenda is here:
>>http://www.bioontology.org/wiki/index.php/Networks
>>together also with instructions on how to get to the Newark Airport
>>Hilton, where our meeting will take place. (Look for Junior Ballroom
>>when you get there.)
>>
>>Presenters have been instructed to keep their remarks to 10 minutes
>>or less, in order to leave ample time for discussion. All slides will
>>be posted to the meeting wiki, and presenters are encouraged to send
>>their slides to me in advance (in longer versions if preferred).
>>
>>All participants should feel free to send me any other supplementary
>>material which they think might be useful for posting to the wiki.
>>
>>With greetings
>>Barry
>>http://ontology.buffalo.edu/smith
>>
>
>
>
>
>>
>
Robert Arp, Ph.D.
National Center for Biomedical Ontology (NCBO)
NYS Center of Excellence in Bioinformatics & Life Sciences
The University at Buffalo
(716) 881-7905 = Office
(850) 509-4503 = Cell
ra...@buffalo.edu
http://www.org.buffalo.edu/rarp/