Search Mailing List Archives
Gene attributes that don't fit into process, component and fu nction ontologies?
froth at hms.harvard.edu
Fri Feb 16 07:46:29 PST 2001
While the ontology at http://www.cbil.upenn.edu/Ontology/MGED_ontology.html
looks at first glance to be an excellent ontology to describe tissue and
cell samples and describe the microarray (and potentially other)
experiments performed on them, it doesn't really address biochemical and
genetic attributes of genes and gene products that are excluded from
process, function and component as in the examples I gave earlier. I am
really not sure how to break this down into separate ontologies, but as a
start maybe: intrinsic properties, regulation, modifications, interactions
(see a rough example below).
We certainly wouldn't want to duplicate the SwissProt effort, but are they
developing an ontology, or a flat set of controlled-vocabulary terms? One
might consider developing an ontology with the same format and using the
same tools as GO, and then map annotation from Swiss-Prot keywords to this
Rough attempt at example ontology extensions:
% intrinsic properties
% contains intron(s)
% differentially spliced
% codon usage
% domain family homology
% regulated degradation
% regulated translation
% DNA level
% transcript level
% methyl G capped
% protein level
% interactions (excluding protein complexes already covered in component
% DNA level
% RNA level
% protein level
At 10:31 AM 2/16/2001 +0000, Genetics wrote:
>I think that I was a bit brusque when I replied to Fritz's mail !
>Richard's mail and that from WinHide show that Fritz has touched a chord.
>1. On the question of expression ontologies there is, of course,
>a working party on this due to report at the MGED meeting in
>2. I am not quite sure what WinHide is trying to attain. Unless I
>have completely missed the point (not unknown, I am told) most
>of his list is satisfied by a well formed GenBank/EMBL-Bank/DDBJ
>FT statement for a mRNA, e.g.
>Win, cannot we compute most of what you want from that ? and then
>represent it in text is you so wish ?
> >The relative expression of each exon
> ... I am not sure what you mean here; do you mean expression of
>each differently spliced protein isoform ? Does it make sense
>to talk about the expression of an exon independently of the mRNA
>within which it is contained ?
>3. With respect to Richard's email (which I take the liberty of
>attaching as it was not sent to the list) I do agree that we must
>extend the domains of cvo's. I also agree that the GO Consortium could
>evolve into an umbrella group for related cvo's (at the very least
>for their distribution, but also I would hope for uniformity of
>syntax, semantics and tools for maintaining and querying). With that
>in mind I would certainly include the following:
>But I do not think that we should attempt to duplicate work being done
>by, eg, the Swiss-Prot team on proteins.
> >From R.BRUSKIEWICH at cgiar.org Fri Feb 16 00:13:58 2001
>Envelope-to: ma11 at gen.cam.ac.uk
>Delivery-date: Fri, 16 Feb 2001 00:13:58 +0000
>Date: Thu, 15 Feb 2001 15:55:31 -0800
>From: "Bruskiewich, Richard" <R.BRUSKIEWICH at CGIAR.ORG>
>Subject: RE: Gene attributes that don't fit into process,
> component and fu nction ontologies?
>To: "'ma11 at gen.cam.ac.uk'" <ma11 at gen.cam.ac.uk>, froth at hms.harvard.edu
>X-Mailer: Internet Mail Service (5.5.2653.19)
>Hi Michael and Fritz,
>I generally agree with you Michael, that we need to keep the current GO
>project tractable by limiting the controlled vocabularies/ontologies (cvo)
>to the three current themes.
>That said, I think we do need to start thinking along the lines Fritz has
>indicated and chart out a broader map for the future. I'm already doing this
>within the plant community by developing trait and mutant descriptive
>ontologies, leveraging the protocols and tools of the GO consortium for
>cvo's. There are also other efforts in biological ontology for example the
>microarray groups, even the various XML driven annotation data exchange
>protocols, all represent cvo's.
>Our recent Plant & Animal Genome workshop on cvo's was well received by
>delegates. Based upon my discussions during a recent trip to Japan, our
>japanese colleagues are also keen to participate. I think many people are
>recognizing the value of community level protocols for labeling database
>data to facilitate tractable functional genomic computing and
>interoperability of heterogeneous databases.
>The basic question is: can we identify a consortium of consortia dealing
>with cvo's in various targeted domains and encourage the development of
>common protocols and tools, leaving content development to each specialist
>International Rice Research Institute (IRRI)
>Tel. +63 (2) 845-0563 Fax +63 (2) 845-0606
>Email: r.bruskiewich at cgiar.org
>Mailing address: MCPO Box 3127, 1271 Makati City, The Philippines
>Shipping: Suite 1009, 6776 Ayala Ave, Makati City, The Philippines
Frederick P. Roth, Assistant Professor
Harvard Medical School
Department of Biological Chemistry and Molecular Pharmacology
240 Longwood Avenue, C-212
Boston, MA 02115
(617) 432-6224 phone
(617) 738-0516 FAX
froth at hms.harvard.edu
This message is from the GOFriends Mailing list. A list of public
announcements and discussion of the Gene Ontology (GO) project.
Problems with the list? E-mail: owner-gofriends at genome.stanford.edu
Subscribing send "subscribe" to gofriends-request at genome.stanford.edu
Unsubscribing send "unsubscribe" to gofriends-request at genome.stanford.edu
More information about the go-friends