See also: IRC log and Vipul's agenda
Holger: I spoke with Chris this morning and will be presenting the slides
<ericP> => http://www.ifomis.org/acgt/coi-presentation.ppt slides
Cris: I will give the presentation. ACGT means advanced clinical genomic trails.
ACGT is an OWL-DL ontology. Formally defined relations and other restrictions.
ACGT leverages BFO (Basic Formal Ontology) as the top-level ontology
<ericP> Chris: adding an OWL anatomy ontology
As of Oct 2007, there were a total of 1109 classes (primitive: 1028, defined: 81)
Cris: there were a total of 121 restrictions
<ericP> Cris: constraints extension includes identifying disjoint classes
Cris: important upcoming work is to start adding FMA (anatomy) classes
<ericP> Helen: will you be able to do classifications with your reasoner after adding FMA?
<ericP> Cris: old slides; we have given up on this because FMA doesn't lend itself well to OWL
<ericP> Helen: try Galen
<ericP> Cris: Alan Rector suggested we set our sights on something more modest
Cris: we spoke with Alan Rector
and he suggested something less complex
... "Patient" is a role in ACGT and Jennifer agrees with that
characterization
ericP: difference between a role and a object?
Cris: role is a dependent
continuant and object is an independent continuant.
... dependent continuant depends on something else for its own
existence. E.g., quality of an "item" depends on the "item" for
its own existence.
... role is a dependent continuant because certain entities can
play different roles, and so it can attach to independent
continuant at different times
<ericP> [slide 26]
Holger: ACGT contains concepts
for clinical trials and genomic research
... concepts/relationships encoded in the ontology are
represented in Trial of Principle (TOP Trial) forms
ericP: we will discuss the last
agenda item
... we will NOT discuss the last agenda item :-)
canceled as Parsa isn't present.
<ericP> holger_stenzhor1: was hoping Christian could address this
Holger: Cris will probably be
able to explain the mapping issue much better. ACGT does not
directly encode the mappings directly.
... ACGT partners (from Greece?) will be addressing the mapping
issue
<ericP> ACTION: Holger to invite Chris to discuss Mappings if any between ACGT ontology and CDISC/HL7 or to provide a narrative himself [recorded in http://www.w3.org/2008/03/11-hcls-minutes.html#action01]
Helen: is ACGT.owl is the latest file?
<holger_stenzhor1> www.ifomis.org/acgt/1.0
Holger: yup its latest
<ericP> => http://www.ifomis.org/acgt/1.0 ACGT ontology
Helen: we should think how the
inclusion/exclusion criteria is not buried under the SPARQL
query
... these criteria are part of the trials protocol itself
ericP: not sure if there are other ways of encoding the inclusion/exclusion apart from it being a part of the query