Menu ▾ ▴

#13 ontology revision, part 2

open
nobody
None
5
2006-06-13
2006-06-13
Kirill Degtyarenko
No

1) As a result of discussion between Marcus, Alan and
myself, we have concluded that we *do not want* to
insist on every term to be linked via is_a relationship
to the one of three main ontologies.

2) We can remove the word "ontology" from the three
main ontologies, and the word "molecular" from
"molecular structure". The redesigned top level looks
like this:
Root ChEBI ontology

%structure (formerly 'molecular structure ontology')
(plural classes)
molecular entities
groups
atoms = elements
particles

(groups, atoms, particles will be linked by
corresponding is_part_of relationships)

%application (formerly 'application ontology')
singular thing e.g. solvent
%biological role (formerly 'biological role ontology')
singular thing e.g. cofactor

3) Ontology export/visualisation.
Desirable: to be able to choose which / how many
lineages to show (e.g. only application; or application
& structure)

4) Also could be useful to be able to see the
unclassified list (internally)

K.

Discussion

  • pmatos

    pmatos - 2006-06-13

    Logged In: YES
    user_id=1169048

    Hi,

    As I joined the discussion late, I have the following
    questions for each of the points
    1) Is this not possible or rather why is this not possible?
    Seems like something very important for some of our users?

    2) Do the three ontologies still have a part_of relationship
    with ChEBI_ontology? Also they wanted us to rename
    ChEBI_ontology to something else like chemicals?

    Are we running the risk that calling the upper ontology
    'structure' that it is too general a word? It will mean
    different things to different people. So shouldn't we rather
    be more specific i.e. chemical structure.

    3) Agree.

    4) Its rather large so I will provide it as a search
    facitlity under the search tab. I can't show it under the
    tree view.

    P.

     

  • Kirill Degtyarenko

    Kirill Degtyarenko - 2006-07-24

    Logged In: YES
    user_id=802919

    Hi,

    it maybe is a hair splitting, but as soon as we had ChEBI-24
    and suggested the new ontology for the next release

    (i.e.
    CHEBI:24431 molecular structure
    CHEBI:24432 biological role
    CHEBI:33232 application
    CHEBI:33233 elementary particle),

    I've noticed that we did not use "elementary" (see above)
    and there was a good reason for that. For lot of people will
    disagree with our def of elementary particles that include
    nucleons. Therefore it could be a better idea to have
    something like "subatomic particle" at the top level with
    "elementary particle (or fundamental particle) is_a
    subatomic particle".

    Happy Big Bang ChEBI

    K!