bpeters42
commented
6 years ago This is beautifully laid out. I understand black boxes to be instances, and grey boxes to be classes. I am a bit unclear about the 'specifies value of' relationship between the 'Complete glabella region' and and the specific instance 'Completeness of glabella 01'. I was also expecting some kind of 'completeness assessment process'. But one problem for me is that I am not sure I completely understand the domain. I would love to talk this through on a call. The example is very much in scope, and it needs to be possible to document it as you have done here.
It might be useful to switch to an example that is easier to grasp for people outside of the domain other than skeletal completeness, but that has the exact same structure that you are looking for. For example, a study using blood samples that were biobanked, and where only samples stored in a certain type of tube are considered as being appropriate for a study.
On Thu, Sep 27, 2018 at 7:40 AM cuboideum notifications@github.com wrote:
@jamesaoverton https://github.com/jamesaoverton has encouraged us https://github.com/RDFBones/RDFBones-O/issues/112#issuecomment-382013841 to contribute a use case involving categorical data specifications. Use Case
We https://github.com/RDFBones are implementing the OBI in a software supporting biological anthropologists with the documentation of their osteological studies on human skeletons from archaeological or forensic backgrounds. Completeness and preservation of these skeletons are recorded in 'skeletal inventories' which we model as data sets. Study designs of various types of investigations formulate requirements (plan specifications) for data items in these datasets that need to be met for a skeleton to be included in a certain investigation. By matching requirements defined in the study design with measurement data of skeletal inventories that are identified as investigation input in the plan concretising the study design for a specific investigation, the software can automatically identify adequate material. This will help so semi-automate the specimen collection process. In case of a match, the considered material can serve as a specimen, otherwise it is not considered.
I assume that in the current OBI version (v2018-08-27) our use case would look like this:
[image: categoricalvaluespecification-obisolution] https://user-images.githubusercontent.com/8103073/46141224-9bfe2b80-c253-11e8-9b80-d736a5192a8c.png
Here, the definition of the categorical value specification 'Complete Glabella Region' would contain the information that measurement data of the type 'Completeness of Glabella' need to have the category label 'complete' in order to establish inclusion in an investigation of type 'FrSexEst'. In the given example, this is not the case but this can only be figured out by reading and understanding the textual definition.
For our ends, we need the categorical value specification to refer to the 'complete' label in order to verify coincidence of value and specification by means of a query. So we would like to do something like this:
[image: categoricalvaluespecification-rdfbonesmismatch] https://user-images.githubusercontent.com/8103073/46141898-93a6f000-c255-11e8-9110-33156ea370a0.png
In contrast to 'Complete Glabella Region', the categorical value specification 'Required glabella region' holds its information not in the textual definition annotation but by denoting the required category label. In another example, where the measurement datum satisfies the specification, a specimen is established:
[image: categoricalvaluespecification-rdfbonesmatch] https://user-images.githubusercontent.com/8103073/46141996-d36dd780-c255-11e8-8a2b-22c6b338f8f4.png
Our software application should be able to process different types of skeletal inventories and different types of study designs. Our queries, therefore, need to be sufficiently generic. Coding the value specification in a textual annotation would require a specific query for each specification. Questions
- Is this a correct usage of categorical value specifications? Are there alternative solutions?
- Is obo:IAO_0000219 ('denotes') the correct property to use?
We would be grateful for feedback on this problem.
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/obi-ontology/obi/issues/968, or mute the thread https://github.com/notifications/unsubscribe-auth/ANN9IoiNiycZgkQFgG52v-fw8Fo3I4VZks5ufLkogaJpZM4W8eD2 .
-- Bjoern Peters Professor La Jolla Institute for Allergy and Immunology 9420 Athena Circle La Jolla, CA 92037, USA Tel: 858/752-6914 Fax: 858/752-6987 http://www.liai.org/pages/faculty-peters
cuboideum
cstoeckert
ddooley
@jamesaoverton has encouraged us to contribute a use case involving categorical data specifications.
Use Case
We are implementing the OBI in a software supporting biological anthropologists with the documentation of their osteological studies on human skeletons from archaeological or forensic backgrounds. Completeness and preservation of these skeletons are recorded in 'skeletal inventories' which we model as data sets. Study designs of various types of investigations formulate requirements (plan specifications) for data items in these datasets that need to be met for a skeleton to be included in a certain investigation. By matching requirements defined in the study design with measurement data of skeletal inventories that are identified as investigation input in the plan concretising the study design for a specific investigation, the software can automatically identify adequate material. This will help so semi-automate the specimen collection process. In case of a match, the considered material can serve as a specimen, otherwise it is not considered.
I assume that in the current OBI version (v2018-08-27) our use case would look like this:
Here, the definition of the categorical value specification 'Complete Glabella Region' would contain the information that measurement data of the type 'Completeness of Glabella' need to have the category label 'complete' in order to establish inclusion in an investigation of type 'FrSexEst'. In the given example, this is not the case but this can only be figured out by reading and understanding the textual definition.
For our ends, we need the categorical value specification to refer to the 'complete' label in order to verify coincidence of value and specification by means of a query. So we would like to do something like this:
In contrast to 'Complete Glabella Region', the categorical value specification 'Required glabella region' holds its information not in the textual definition annotation but by denoting the required category label. In another example, where the measurement datum satisfies the specification, a specimen is established:
Our software application should be able to process different types of skeletal inventories and different types of study designs. Our queries, therefore, need to be sufficiently generic. Coding the value specification in a textual annotation would require a specific query for each specification.
Questions
We would be grateful for feedback on this problem.