n-ary synteny blocks, and a model question

classic Classic list List threaded Threaded
1 message Options
Reply | Threaded
Open this post in threaded view
|

n-ary synteny blocks, and a model question

Joel Richardson

On the dev call yesterday, it was suggested that synteny blocks
be n-ary rather than binary. What about something like the following?

<class name="SyntenicRegion" extends="SequenceFeature" is-interface="true">
   <reference name="block" referenced-type="SyntenyBlock"
reverse-reference="regions" />
   </class>

<class name="SyntenyBlock" is-interface="true">
   <attribute name="name" type="java.lang.String" />
   <collection name="regions" referenced-type="SyntenicRegion"
reverse-reference="block" />
   <collection name="dataSets" referenced-type="DataSet"/>
   </class>


I'd rather have SyntenyBlock be a subclass of BioEntity (to inherit
name, id, the ability to have OntologyAnnotations, etc.) Unfortunately,
BioEntity is organism specific, while a SyntenyBlock is multiorganism.

I'm wondering if in the core model it would be feasible to split out
an abstract superclass, e.g. Entity, containing:
        - primaryIdentifier
        - name
        - ontologyAnnotations
        - dataSets
        - synonyms
        - publications
        - crossReferences
Then BioEntity would extend this class and add:
        - symbol
        - secondaryIdentifier
        - organism
        - locations
        - locatedFeatures
The idea being that there are cases (and I've already run into two)
where the objects have names, ids, etcs, maybe have annotations of
one sort or another, belong to data sets, etc., but are not necessarily
"bio".

Cheers,
Joel






--

===============================================================
Joel Richardson, Ph.D.
Sr. Research Scientist
Mouse Genome Informatics
The Jackson Laboratory   Phone: (207) 288-6435
600 Main Street          Fax:   (207) 288-6132
Bar Harbor, Maine 04609  URL:   www.informatics.jax.org
===============================================================

_______________________________________________
dev mailing list
[hidden email]
http://mail.intermine.org/cgi-bin/mailman/listinfo/dev