a few questionable superclasses (brain)

[Alan Ruttenberg]

Medial sulcus,Sulcus limitans, and Rhomboid fossa of fourth ventricle
had two parents (one is-a, one metaclass?), one of which is Region of
wall of fourth ventricle (material) and one of which is 'Subdivision
of cavity of fourth ventricle' (immaterial) but immaterial is disjoint
from material.

White matter of left hippocampus had White matter of right hippocampus
as superclass or
metaclass - right/left at the same time conflict.

[mej...@comcast.net]

Hi Alan,

Just came back from a R&R. I'll work on all the issues you raised in the next few days.

Thanks!
Onard

[mej...@comcast.net]

One of the problems with using Protege frames is when moving classes from one regular superclass to another regular superclass, the corresponding standard class or metaclass assignment is not automatically implemented. Such is the case below when we decided that medial sulcus and sulcus limitans are not material objects but rather spaces and therefore moved them from "region of wall of fourth ventricle' to 'subdivision of cavity of fourth ventricle' but forgot to re-assign their metaclasses. There are probably quite a few of these existing and if you can programmatically generate a list of these please send it to me. I'd appreciate that very much.

WRT to white matter of left hippocampus as child of white matter of right hippocampus, the current copy of the FMA does not have that conflict.

Thanks again!

[Alan Ruttenberg]

On Mon, Jun 7, 2010 at 4:28 PM, <mej...@comcast.net> wrote:

One of the problems with using Protege frames is when moving classes from one regular superclass to another regular superclass, the corresponding standard class or metaclass assignment is not automatically implemented. Such is the case below when we decided that medial sulcus and sulcus limitans are not material objects but rather spaces and therefore moved them from "region of wall of fourth ventricle' to 'subdivision of cavity of fourth ventricle' but forgot to re-assign their metaclasses. There are probably quite a few of these existing and if you can programmatically generate a list of these please send it to me. I'd appreciate that very much.

How do I know which they are? Is any case where the metaclass differs from the superclass an error?

 

WRT to white matter of left hippocampus as child of white matter of right hippocampus, the current copy of the FMA does not have that conflict.

Hmm. I still see this - perhaps one of them is hiding in the metaclass (as I remove the distinction in my version). I can check again when I download the next version.

FMA::|White matter of left hippocampus|

  :IS-A (FMA::|White matter of right hippocampus|

         FMA::|White matter of hippocampus|)

  :SLOTS ((FMA::FMAID "275171")

          (FMA::|Preferred name| FMA::|ONARD_Instance_2820125|)

          (FMA::|regional part| FMA::|Fimbria of left hippocampus|))

  :TYPE :CLASS

-Alan

[mej...@comcast.net]

Where the metaclass differs from the superclass is NOT necessarily an error.  There are very few cases where the class has itself as its metaclass for the purpose of propagating its direct attributes not only to its subclasses but to itself as well. I can't think of a specific example right now but I'll get back to this as soon as I find an example.

OM