Search Mailing List Archives


Limit search to: Subject & Body Subject Author
Sort by: Reverse Sort
Limit to: All This Week Last Week This Month Last Month
Select Date Range     through    

Fwd: Re: linking the function to the process

Fritz Roth froth at hms.harvard.edu
Thu Jun 26 11:48:35 PDT 2003


I hasten to add that the GO Bayesian network DAG that we produced no longer 
follows Judy's true path rule.  In our GO Bayesian network, edges represent 
dependency relationships that need not be absolute (i.e., logical) 
dependencies.  However, strong dependencies could be reviewed by someone 
who knows the biology to decide whether each new dependency between (or 
within) the GO DAG does actually follow the true path rule.  Any 
volunteers?  :)

Cheers,
Fritz


>Re: relating GO terms to each other "Kinda like: users who bought this Go 
>term also bought this one", as mentioned by Evelyn Camon...
>
>You may want to check out our paper predicting GO terms based on patterns 
>of other GO terms 
>(http://llama.med.harvard.edu/publications/King_GenRes_2003.pdf ).  We 
>draw the very clear analogy between our approach and the "collaborative 
>filtering" methods used by Amazon.com and others to make product 
>recommendations to customers.  For this project we built a Bayesian 
>network of GO terms, a DAG that includes the three GO DAGs as well as 
>other connections learned from current SGD and FlyBase GO annotation, 
>including links between terms in different GO DAG branches.
>
>We also have a related paper in press at ISMB relating GO terms to 
>phenotypes ( 
>http://www.iscb.org/ismb2003/paperAbstracts/btg1024.pdf  ).  As Arne may 
>know (on not, Aventis is a big company!), both of these projects were 
>sponsored by Aventis.
>
>Cheers,
>Fritz Roth
>
>
>
>At 05:38 PM 6/26/2003 +0100, Evelyn Camon wrote:
>>Hi,
>>
>>In the QuickGO browser at the EBI we have a facility that
>>suggests which GO terms across all the ontologies are annotated in tandem
>>to another GO term. The concurrent assignment is based on real 
>>annotations in the
>>database.
>>I think AmiGO are also going to implement such a system using all the GO
>>Consortium annotations. Kinda like: users who bought this Go term also bought
>>this one?
>>eg. DNA binding and nucleus.
>>There is no formal link between the ontologies but I agree it would be useful
>>sometimes.
>>
>>kind regards
>>Evelyn Camon
>>(EBI Swiss-Prot GO Annotaion project)
>>
>>Arne.Muller at aventis.com wrote:
>>
>> > Hello,
>> >
>> > I guess in theory biological process, molecular function and celluar
>> > comartment are disjoint entities, they should be independant. In 
>> practive off
>> > course they are not. I think I'd be a good idea to develop such a link
>> > between the three catgories.
>> >
>> > I think at the moment you cannot link the function to the process, but 
>> if you
>> > have a gene product you could just get the function AND the process.
>> >
>> >         Arne
>> >
>> > >
>> > > Hi,
>> > >
>> > > Given a function, how do I find out what biological processes it is
>> > > associated to?  I tried GO ID or key word descriptions of
>> > > this function on
>> > > those web tools listed on GO website and they didn't seem to work.
>> > >
>> > > Thanks.
>> > >
>> > > Ronglin Wang
>> > > Syngenta
>> > >
>> > >
>> > > --
>> > > 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/
>> > >
>> > >
>> >
>> > --
>> > 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/
>>
>>
>>--
>>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/

_________________________________________________
Frederick P. Roth, Asst. Professor
Harvard Medical School
Dept. of Biological Chemistry and Molecular Pharmacology
250 Longwood Avenue, SGMB-322, Boston, MA 02115
(617) 432-3551 phone            (617) 432-3557 FAX
froth at hms.harvard.edu           http://llama.med.harvard.edu
_________________________________________________


--
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/



More information about the go-friends mailing list