Re: NeuroLex help

[anita bandrowski]

Hi Maryann,

There are two issues that came up in conversation today with Mihail that concern the greater NeuroLex list.

It would be great if we can get some ideas about the validity of the overlaps relationship when discussing the platonic brain regions such as basal ganglia.

Since the regions are by definition not tied to a species or space, Mihail brought up some concerns about using a spatial relationship.

There is also a question about the Pallidum, and how distinct it is from the Globus pallidus. NeuroLex had a term Pallidum, but this was deleted several years ago, and pallidum was made a synonym of globus pallidus. However, Mihail believes that the Pallidum should have part Globus Pallidus because in most definitions there are more regions in the pallidum than in the globus pallidus. I have changed the Globus Pallidus page to reflect this view, but want to throw this out to the community, just in case there is a compelling reason to keep them synonymous.

Regards,
anita

On Tue, Nov 26, 2013 at 4:05 PM, Mihail Bota <mbo...@gmail.com> wrote:

Hi Maryann,

I think that a rephrase of the paragraph from the Swanson 1998 Atlas
would be: Basal nuclei are made of those telencephalic (endbrain)
regions that develop from ventricular regions". I used "telencephalon"
because I've seen this term used in several concepts. The criteria for
defining Basal Nuclei would be ontogenetic, then
topographical/observation and then cytoarchitectural, which is an
instantiation of the chemo criterion.

I don't think I understood what "related" stand for, this is why I've
inserted those terms.

Aggregate: I think all NS regions are aggregates. One could sub-divide
regions up to the level of neuron types. I usually take the top-down
approach of defining regions, and I've seen that other terms use the
same approach. On the other hand the "aggregate" field is very
powerful if one includes the passing fibers, dendrites, local axons...

Yes, please change what is needed to be changed, to conform to the
NeuroLex syntax and rules.

As a general question: what do you think of a database-driven (or RDF)
version of NeuroLex? I noticed the defining criteria fields that
Stephen recently added. Each of these criteria is composed of other
sub-criteria, and variables, which actually are the properties of
neurons and regions that are observed and measured. I'm also thinking
that the connections are in forms of lists...

Mihai

On Tue, Nov 26, 2013 at 5:33 AM, Maryann Martone <mar...@ncmir.ucsd.edu> wrote:
> Thanks Mihail.  I looked these over.  For the definition field, we try to stick with Aristotelian-type definitions, i.e., A is a type of B that has C, although we don't always strictly adhere to this when it doesn't make sense.  So for these types of aggregate structures, I will often say "Collection of brain regions"…   I couldnt quite tell from the definition which parts of the brain were being included in this definition.  I'm assuming the striatum and palladium.   So I would say that striatum and pallidum are part of basal nuclei rather than related to.  Also, since Basal nuclei, in this case, comes from Swanson, we would label this "Basal nuclei of Swanson 1998" and label it a brain parcel.  Is it OK if I take a pass at modification and you see whether it captures your intent?
>
> Regards,
>
> Maryann
>
>
> Maryann Martone, Ph. D.
> Professor-in-Residence
> Department of Neuroscience
> University of California, San Diego
> San Diego, CA 92093-0608
>
> 858-822-0745
>
>
>
>
> On Nov 25, 2013, at 5:33 PM, Mihail Bota <mbo...@gmail.com> wrote:
>
>> Hi Anita,
>>
>> I started 2 pages in NeuroLex, Basal Ganglia (mammals) and Basal
>> Nuclei (rat), and I realized that I'd need a bit of help with the
>> interface.
>> May I ask you to look over my activity in NeuroLex (user mihai10), to
>> see whether I've inserted correctly so far? Also, how can I link a
>> generic term to a more specific one, species-wise?
>>
>> Thank you,
>>
>> M.
>>
>

--

Anita Bandrowski, Ph.D.
NIF Project Lead
UCSD 858-822-3629
http://neuinfo.org
9500 Gillman Dr.#0446
la Jolla, CA 92093-0608

[Chris Mungall]

I think it's useful to look at how this has been modeled in different ontologies. Uberon follows the FMA in distinguishing the basal ganglia (qua aggregate) and an individual member of this aggregate. This may not be the best way to model these structures, but it makes the use of spatial relations clearer. For example, being contained in an aggregate does not entail being contained in any member of that aggregate.

Note that it's still valid to use an overlaps relation (or any other region relation) with generic aggregates (assuming RCC8 in OWL semantics, not clear if Neurolex commits to this).

I'm not qualified to offer advice on globus pallidus vs pallidum distinctions, but I note that in ontologies such as FMA and MA these are considered distinct.

--
You received this message because you are subscribed to the Google Groups "neurolex" group.
To unsubscribe from this group and stop receiving emails from it, send an email to neurolex+u...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[Mihail Bota]

Hello,

As discussed previously with Maryann and today with Anita & Stephen,
the basal ganglia (BG) is actually defined functionally, or
clinically. What was initially defined based on observation on the
tissue was then re-defined as a set of regions that are involved in
motor sequences. If one looks in a neuroanatomical atlas of the rat
will see that the caudoputamen and globus pallidus are much more
rostral than the subthalamic nucleus and substantia nigra, and the
first 2 regions are not tangent with the latter ones. This would be
the first and the BG-specific counter-argument for aggregate.
The second counter-argument is that any region of the brain can be
modeled as an aggregate, but the truth is that brain regions are
representations. Some are defined by how the neurons look and are
arranged, other by the fact that stimulation elicits a motor response,
etc. If one uses the aggregate relationship, then the smallest regions
would be those made of specific neuron classes or types. So, in my
opinion the aggregate relationship gives a contradiction when the
entire chain is followed top-down and then bottom-up.
The third counter-argument is specific to the science we try to
formalize in ontologies. Neuroanatomy uses topographical relations
(ventral, dorsal, etc.), and topological ones (X is included in Y; A
is bordered by B). Hence, I think we encode for these relations...
However, I don't think that FMA should change the relationship, if it
helps it in organization and inferences.

The "overlap" relation is very important when comparing two atlases of
the brain in the same species. However, I don't see how it can be used
when we're talking about two different species, because we're in the
situation of comparing countries from two different planets.
As far I know, there are three (related) ways to compare regions
across species. The first and strongest is homology, but to be sure
that A is homologous with B we need to know how they looked in an
intermediary species. Thus, I'd relax this comparison with "structural
similarity", which has the advantage that it can be appended with
structural attributes (common cell types, gene expression patterns,
neighbors, etc.). The second is the part of the embryonic tissues that
gave rise to the compared regions. And the third is the classic one,
not much used today, that of representing A and B onto a generic
structural plan, the Bauplan.

Yes, I think that Pallidum should be a superpart of the Globus Pallidus, too.

Mihai