[Go-database] terms to which a gene is not annotated

[Chris Mungall]

On Mon, 27 Sep 2004, Steve Carroll wrote:

> Chris,

Hi Steve

> Thanks, that helped a lot.  I actually realized that genes can be
> annotated to several terms (i.e. have several roles), I should have worded
> my question on that aspect better.  What I meant was whether there exist
> ANY two terms in the GO that are mutually exclusive.  I gather that if
> this were the case, and a gene were annotated to one of them, it would be
> "negatively annotated" (i.e. with the NOT qualifier, which I learned about
> today) to the other.

Hmmm, this still isn't the case - see my example about nucleus and
extracellular space. These are mutually exclusive in the sense that an
entity cannot be in both _at the same time_ - yet annotation to one does
not necessarily exclude the other

Hope that clears it up
Chris

> Thanks again,
> Steve
>
>
> On Mon, 27 Sep 2004, Chris Mungall wrote:
>
> >
> >
> > On Mon, 27 Sep 2004, Steve Carroll wrote:
> >
> > > Hello,
> > >
> > > I am a graduate student doing a project that involves the GO, and I have
> > > what may be a naive question:  for a given gene, there is a list of terms
> > > to which that gene is annotated, so with some confidence we can say that
> > > gene belongs to those classes (and their ancestors in the GO).  But can we
> > > also say that the gene in question does NOT belong to the classes defined
> > > by the terms to which it is not annotated?  Or is it better to treat
> > > membership in those classes as unknowns?  Does the answer to this question
> > > depend on the gene or the actual terms to which it is annotated?  For
> > > example, does annotation to some term preclude the possibility of
> > > annotation to some other term?
> >
> > Hi Steve
> >
> > In general, the possibility of annotation to another class is never
> > precluded by annotation to a different class
> >
> > Even when we consider two classes that are disjoint - for example, the
> > nucleus does not border or overlap the extracellular space - it is still
> > possible for a gene product to be active in the nucleus in one context and
> > for the same gene product to be active in the extracellular space in some
> > other context. Gene products frequently have multiple roles.
> >
> > The only time we can say with confidence that a gene product is not
> > associated with a class (or the *descendants* of that class) is when a
> > curator specifically makes a negative association, based on experimental
> > evidence and expert judgement. See the association format docs, or the go
> > database docs if you wish to query based on these.
> >
> > Cheers
> > Chris
> >
> > > In case it is of any relevance, I am using the SGD database for my
> > > application.
> > > I am not subscribed to these lists, so please respond directly to me.
> > > Thankyou in advance for any help.
> > >
> > > Sincerely,
> > >
> > > Steve Carroll
> > > M.S. Student
> > > Computer Science
> > > Rutgers University
> > >
> > > _______________________________________________
> > > Go-database mailing list
> > > Go-database at mail.fruitfly.org
> > > http://mail.fruitfly.org/mailman/listinfo/go-database
> > >
> >
>
>

--
This message is from the GOFriends moderated mailing list.  A list of public
announcements and discussion of the Gene Ontology (GO) project.
Problems with the list?           E-mail: owner-gofriends at geneontology.org
Subscribing   send   "subscribe"   to   gofriends-request at geneontology.org
Unsubscribing send   "unsubscribe"  to  gofriends-request at geneontology.org
Web:          http://www.geneontology.org/