[223]                             home                             [225]

 

Sunday, June 11, 2006

 

Challenge problem à

The Taos Discussion à

 

Generative Methodology Glass Bead Games

 

 

On ontological modeling of expression

 

The metaphor between gene, cell and social expression  à [217]

On Formal verses Natural systems à [206]

 

 

 

To: biopax-discuss@cbio.mskcc.org

Subject: [BioPAX-discuss] Three proposals for BioPAX

 

 

 

I have collected three proposals in a single document (.pdf attached).

 

The proposals are:

* Enhancing BioPAX with SKOS (to be integrated into the URI/CV proposal of Alan and Andrea... sorry that I am late)

* Removing PhysicalEntityParticipant

* Renaming ‚utilityClass' and ‚externalReferenceUtilityClass'

 

kind regards,

Matthias Samwald

 

 

Response from BCNGroup..

 

 

I would like to ask a simple question.

 

In the W3C specs at:

 

http://www.w3.org/2004/02/skos/extensions/spec/

 

it is specified that a "definition" is

 

   "An explanation of the meaning of a class or property."

 

 

There is a sizable part of the theoretical foundations to biology that real biological processes produce "systems" whose function is determined by the environment.

 

Robert Rosen might go further and suggest that the interactions between PhysicalEnitityParticipants expresses both some selection of endophysical causality and expression of some exophysical causality (using Otto Rossler's terms) - both types which emerge in the situation and not before.  If this is true, the real meaning of a reaction leading to a signal is rooted in a pragmatic axis, not solely in the concept called "semantics", as in the "Semantic Web".  Yes, semantics is related to the formation of natural category and the re-expression of natural category....  but is semantics the final cause of "meaning? 

 

Can the meaning of a physicalEntityParticipant be determined uniquely and well specified as a URL?

 

QSAR, qualitative structure relationship analysis, is a non statistical means to anticipation state transitions and "signal expression".  But QSAR is not popular within the mainstream funded science.  What is popular is reductionism. 

 

If this theoretical issue is set aside, one can in fact define uniquely the meaning of things as a set of URLs, and let the definition of "semantics" be full specified (so that for example one of the Description Logic engines - Racer, for example - can make "inferences").  This specification can also be "owned", and as a result the body of scientific knowledge about plants and biochemistry will move from being part of the cultural heritage to things owned by individuals. 

 

But should the biologist consent to this theoretical issue being set aside?  Two parts to the question. 

 

One: is it proper science?

Two: are we, as natural scientists, aware of the consequences of making specifications property?

 

If you say, "no, of course not.  The issue is a valid issue."; then how does one square the deeply ontological claims that class-subclass assertions make on how human organize scientific knowledge? 

 

If it is a valid issue, then does non-locality play a role in the emergence of function, and thus the "meaning of a signal"?

 

How can one use classes and subclass hierarchies to well specify non-locality such as field properties?

 

 

 

Dr Paul S Prueitt

Taos New Mexico