Mammalian brain anatomy
From Program on Ontologies of Neural Structures
One of the tasks of the PONS task forces is to develop a Structural lexicon of pan-mammalian structures.
This will be the focus of the PONS Seattle Summer 2010 meeting. In preparation for that, Stephen Larson has prepared a draft spreadshee attempting to lay out brain regions across species. Please note that this is an early draft and there are as yet unaddressed issues that have been raised.
Stephen's spreadsheet
Stephen's notes:
What's included
- Worksheets for NeuroNames general, Hof, Swanson 04, Swanson 98, Dong, Paxinos & Franklin Rat, Paxinos Macaque
- NeuroLex labels and IDs generated. Labels are based on a concatenation of the original name and the nomenclature reference. IDs have been turned into an nlx_ code and a counter
- Nomenclature and species columns
What's missing:
- A super category column for every worksheet that relates the structure back to NeuroNames. This requires two things:
- Mappings from each atlas into NeuroNames. We appear to have enough information to do this for the Hof, Swanson04, Dong and Paxinos & Franklin atlases. We don't yet have such mappings for Paxinos Macaque, or Swanson 09 rat.
- A bit of excel formula magic to pull data between the worksheets to populate the super category column with names from the neuronames hierarchy.
Doug Bowden's revisions of August, 2010
I have revised the NeuroLex Mammalian Brain hierarchy to address issues raised at the meeting last month. You can see the revised version by following the trail:
- http://braininfo.org
- Search by Name
- brain
- brain
- What Where How Big is It?
- Models Where It Appears
- NeuroLex Mammalian Brain
- Expand All
The changes include:
1- NeuroLex standard name and NeuroNames default name ‘isocortex’ has been changed to ‘neocortex’ on basis that 1) 'neocortex' appears 10 times as often in PubMed, 2) several authorities agree that it is uniquely a synonym of isocortex (not a homonym for ‘true isocortex’) so is unambiguous, and 3) the other parts of Stephan and Zilles's nomenclature retained the paleo- and archi- terminology, so the argument that ‘neocortex’ is not evolutionarily correct has not been applied equally to those concepts. Thus, to return to ‘neocortex’ as the standard term does not violate the criterion of internal consistency of current terminology.
2- Moved ‘proisocortex’ from siblinghood with ‘neocortex’ to child of ‘neocortex’ to be consistent with the most authoritative classifications, Stephan (1975) and Zilles (2004). This corrects an error in our initial posting.
3- Restored ‘true isocortex’ as sibling of ‘proisocortex’ (child of neocortex): 1) so as not to omit a well recognized entity and 2) otherwise neocortex would be a parent with only one child… a no-no for any hierarchical ontology.
4- Deleted three concepts (nodes) for not having been mentioned in more than 1 PubMed abstract in the last 25 years: Allocortex bulbi olfactorii, Allocortex primitivus, and periarchicortex (These will continue to be defined in NeuroNames, but will not appear in the NeuroLex hierarchy.)
5- Continued to identify an inconsistency between your desire to include certain lower level structures in the NeuroLex Mammalian Brain when all species don’t necessarily have all the kids of a given parent. You want the archicortex, because that’s where the hippocampal formation is located, but archicortex also contains the paraterminal gyrus, and fasciolar gyrus, which are present in primates and may not have equivalent structures in the rodent… I don’t know. This may or may not be a problem depending on what you expect the hierarchy to be used for. I don’t see it as a problem for an atlas to be used for quantitative neuroanatomy. It is not good for teaching comparative anatomy, because it implies that all species have all structures included in the hierarchy. I don’t know how it is for a descriptive logic ontology, but I suspect it’s not good. In any case, I have left the fasciolar and paraterminal gyri in there.
6- Changed the BrainInfo default names for several structures to keep BrainInfo’s default names consistent with NeuroLex nomenclature, which reflects the currently accepted architectonic parcellation of cerebral cortex by Stephan, who expressed it in the 1970s and Zilles who has integrated it into leading English and German textbooks of the past two decades , e.g., ‘paleocortex’ for ‘paleocortex (Stephan)’; this entailed changing ‘paleocortex’ to ‘paleocortex (obsolete)’. Same for ‘allocortex’, ‘allocortex (Stephan)’ and ‘allocortex (obsolete)’. Both were necessitated by the gradual acceptance over the last 40 years of Stephan’s 1970s definition of cortex as including the olfactory bulb in place of the earlier, now ‘obsolete’ concept of olfactory bulb as a subcortical structure. Other BrainInfo default name changes: ‘hippocampus’ to ‘CA fields’; Peripalaeocortex’ to ’peripaleocortex’; ‘parasubicular area’ to ‘parasubiculum’; ‘perirhinal area 35’ to ‘perirhinal area’; ‘retrosplenial region’ to ‘retrosplenial area’.
7- Changed all of the NeuroNames definitions of architectonically defined cortical areas to be consistent with the changes described above.
8- Based on interchanges with you, Clif and others, changed both NN default and NeuroLex standard terms as follows: ‘medial preoptic area’ for ‘medial preoptic nucleus’; ‘lateral preoptic area’ for ‘lateral preoptic nucleus’; ‘CA fields’ for ‘hippocampus’; ‘crus cerebri’ for ‘crus of the cerebral peduncle’
9- Removed all children of the lateral, third and fourth ventricles.
10- I’ve tried to make all of the terms in NeuroLex Mammalian Brain and the default terms in NeuroNames/BrainInfo the same with one exception: ‘subgenual area’ in NeuroLex is ‘area 25 of Brodmann (human) in NN’. Our convention for default naming cortical areas of specific authors is such that I couldn’t change it to ‘subgenual area’. In NN ‘subgenual area’ is a synonym of ‘area 25 of Brodmann (human)’ in NeuroNames. I presume that in NeuroLex ‘area 25 of Brodmann (human)’ will be a synonym of subgenual area.
