[Go] Gene Symbols as GO synonyms
Jim Hu
jimhu at tamu.edu
Thu Apr 3 07:55:50 PDT 2008
Hi,
I agree with Judy that gene symbols are not synonyms for GO terms.
After all, the whole point of GO is that each gene should have three
or more GO terms, and each GO term is associated with many genes
across lots of organisms. That's just not what "synonym" means to me.
On the other hand, I understand and sympathize with Jen's explanation
for why the group wanted p53 in the term record. Now that the
associations are there, you can just look for the gene name in the
gene associations, but while since the annotation process is not
complete, there are other situations where finding terms is helped by
having some gene names "associated" with a term at a level that would
not be accepted as a completed annotation - i.e. no evidence code or
reference. I think the gene names here are still potentially useful
in that way; not as true synonyms, but as listings of examples to
guide curation. I'm wondering if something like that would still be
useful to have in the term record.
Not sure what to call it or how to implement, though. Example is too
broad. I'm thinking that you don't just want examples, you want the
cases where something is a paradigm for that GO term. For example,
for GO:0030983:mismatched DNA binding, mutS would be a paradigm
example, along with the MSH _family_ of proteins. But listing every
MSH gene from every organism would be overkill. For broader things
like ATPase activity, you wouldn't want to pick a particular example
as a paradigm, as that molecular function is connected to too many
disparate things.
I don't think RELATED captures this notion. I'd almost prefer to see
just freetext comments, as I suspect that the kind of knowledge from
fuzzy examples I'm talking about hre is not a good ontological idea.
Jim
On Apr 3, 2008, at 5:46 AM, Judith Blake wrote:
> Hi Jen,
>
> When I search GO terms for 'p53', this does not come back as a
> synonym, although synonyms with the string 'p53' as part of the long
> series of strings as a synonym do come back. So I don't think that
> argument holds anymore. I agree that grinding through looking for
> synonyms that are gene symbols is not a productive use of time, but
> such synonyms, I think, should be removed as a matter of course
> while working on the ontology in general as they are found. I do
> not see, if 'p53' is not an issue, where anything else would be an
> issue.
>
> Does anyone object to having gene symbols routinely removed as found
> as synonyms? Certainly the ontology editors can bring such synonyms
> forward if they think there might be an issue. what do you think Jen?
>
> Judy
>
>
> Jennifer Deegan (nee Clark) wrote:
>> Hi Judy,
>>
>> We did think about this, but I remember that the group who work on
>> p53 were very keen that a term covering an activity of this gene
>> product should carry a synonym 'p53' to help them find the term. I
>> am reluctant to remove all gene symbols as synonyms because of
>> that, but thought I would just tidy up these two problematic terms.
>> We could certainly consider removing all gene symbol synonyms in
>> the long term, but it seems to me that the effort required to track
>> them all down and discuss individually whether they are required
>> might not be worthwhile, since we have other time-consuming
>> projects that are a higher priority. Would you agree?
>>
>> Thanks for thinking about it.
>>
>> Jen
>>
>>
>> Judith Blake wrote:
>>
>>> not the least because they aren't synonyms....
>>>
>>> judy
>>>
>>> Jim Hu wrote:
>>>
>>>> I agree with Judy that removing them from all terms is a good
>>>> idea. If us microbiologists ever get our act together and start
>>>> getting all the genes from metagenomics, you'll be glad you did! ;)
>>>>
>>>> Jim
>>>>
>>>>
>>>> On Apr 2, 2008, at 6:06 PM, Judith Blake wrote:
>>>>
>>>>> I agree with your proposal.
>>>>>
>>>>> Would you not remove gene symbols as synonyms from all terms?
>>>>>
>>>>> Judy
>>>>>
>>>>> Jennifer Deegan (nee Clark) wrote:
>>>>>
>>>>>> Hi,
>>>>>>
>>>>>> A couple of GO terms currently contain a problematically large
>>>>>> number of
>>>>>> gene symbols
>>>>>> as related synonyms.
>>>>>> The synonyms were imported into the OBO file during updates
>>>>>> from Enzyme
>>>>>> Commission data.
>>>>>>
>>>>>> The synonyms are a problem because they mess up the display in GO
>>>>>> browsers. For example:
>>>>>>
>>>>>> http://amigo.geneontology.org/cgi-bin/amigo/term-details.cgi?term=GO:0004714&session_id=6392amigo1207145654
>>>>>> <http://amigo.geneontology.org/cgi-bin/amigo/term-details.cgi?term=GO:0004714&session_id=6392amigo1207145654
>>>>>> >
>>>>>>
>>>>>> They are also confusing for users as they resemble annotations,
>>>>>> but are
>>>>>> only related synonyms.
>>>>>>
>>>>>> To resolve this problem, on 16th April we are going to remove
>>>>>> the list
>>>>>> of gene products as related synonyms from the following terms:
>>>>>>
>>>>>> GO:0004714 (transmembrane receptor protein tyrosine kinase
>>>>>> activity)
>>>>>> (201 gene symbols included as synonyms)
>>>>>>
>>>>>> GO:0004715 (non-membrane spanning protein tyrosine kinase
>>>>>> activity) (84
>>>>>> gene symbols included as synonyms)
>>>>>>
>>>>>> If you have any objection to this removal please reply to
>>>>>> Jennifer
>>>>>> Deegan and Emily Dimmer.
>>>>>>
>>>>>> Thanks,
>>>>>>
>>>>>> Jennifer and Emily.
>>>>>>
>>>>>>
>>>>>>
>>>>>
>>>>>
>>>>> _______________________________________________
>>>>> Go mailing list
>>>>> Go at geneontology.org <mailto:Go at geneontology.org>
>>>>> http://fafner.stanford.edu/mailman/listinfo/go
>>>>
>>>>
>>>> =====================================
>>>>
>>>> Jim Hu
>>>>
>>>> Associate Professor
>>>>
>>>> Dept. of Biochemistry and Biophysics
>>>>
>>>> 2128 TAMU
>>>>
>>>> Texas A&M Univ.
>>>>
>>>> College Station, TX 77843-2128
>>>>
>>>> 979-862-4054
>>>>
>>>>
>>>>
>>>
>>
>
>
=====================================
Jim Hu
Associate Professor
Dept. of Biochemistry and Biophysics
2128 TAMU
Texas A&M Univ.
College Station, TX 77843-2128
979-862-4054
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://fafner.stanford.edu/pipermail/go/attachments/20080403/e911f7d6/attachment-0001.html
More information about the Go
mailing list