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    

Semantic Similarity on Genes via Gene Ontology: Relatedness or complementarity?

Phillip Lord phillip.lord at newcastle.ac.uk
Fri Jul 14 04:40:00 PDT 2006


>>>>> "GF" == George Fou <gio.fou at gmail.com> writes:

  GF> I tried to find similarities among genes exploiting knowledge
  GF> derived
  >> From Gene Ontology. More specifically, I tried to fing semantic
  GF> similarity between genes based on the terms that those genes are
  GF> annotated to. Could anyone more connoisseur on this subject tell
  GF> me if this way gives us highly relative genes or complementary
  GF> ones?


I am not sure what you mean by either "highly relative" or
"complementary" genes. But even though I don't understand your
question I will take a crack at answering it. Such hubris!


I think that the short answer is that semantic similarity measurements
will return genes which are closely related in knowledge space.  This
is a different space from others which are used for measuring gene
relatedness -- in general, this means sequence space. We know that
sequence space and knowledge space are correlated with each other; two
genes which have very similar sequences are likely to be annotated the
same way, while two genes with different sequences are likely to be
annotated differently. But this is not absolute and for any two genes
the correlation is not that predictive. For example, GO is quite
shallow over proteins which people consider to be dull (I seem to
remember that lipoproteins fall into this category). So for this sort
of protein you are likely to see close relationships in knowledge
space but distance ones in sequence space. For proteins which
everybody is interested in (receptors for instance) the opposite will
be true. 

To some extent this is a good thing -- Go represents the intuition of
the biological community. Or the prejudice and bias of the community
considered in a different way. 

Phil

--
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