Saturday, January 07, 2006
New
discussion about signal pathways
and complex ontology
[342] ß parallel discussion in “national debate” bead thread
This is part of a discussion that will be moved to a Wiki page soon.
In modeling the data and the biological expression one is faced with taking a union, of sorts, from bioinformatics data collection and archiving and the scientific thought by a community of natural scientists.
This community knows that the whole is often greater than the sum of the parts, and that the holism observed in cell and gene expression varies even if the constituents of the expression are the same.
So if we ask the question of category, we may have different answers depending on the level of observation and the environmental conditions that leads to a specific set of participants.
See a previous discussion about the OWL data ontology being developed by the BioPAX working group. The BioPAX Wiki home page is:
http://biopaxwiki.org/cgi-bin/moin.cgi/FrontPage
Thus the two different requirements are
1) data integration from databases – most of which are about structural information
2) a controlled vocabulary with those terms that have multiple meanings marked as such.
The data integration is to be done with an OWL ontology: the current version which is downloadable from
http://www.biopax.org/release/biopax-level1.owl
The controlled vocabulary is not organized into classes and equipped with any form of logics. However, programs can be written, in Python for example, that use the controlled vocabulary as well as the OWL file (and embedded logics) to produce knowledge sharing and machine pattern recognition analysis as a means to further scientific works.