[Fwd: Re: [annotation] RNA trimethylguanosine cap binding query]

Alexander Diehl adiehl at informatics.jax.org
Thu Jan 17 02:56:16 PST 2008


Harold,

It is true I have made a number of annotations to the term 'GPI anchor 
binding' for CD24a and Thy1.  I was always a bit uncomfortable about 
these annotations as they violated the idea of binding terms being for 
non-covalent bonds only, but in fact I was only following the precedent 
set by others in the GO for GPI-anchored proteins.  You see for as long 
as I have been in the GO, and perhaps for much longer, we have had the 
SwissProt Keyword mapping,

SP_KW:KW-0336 GPI-anchor > GO:GPI anchor binding ; GO:0048503

which basically means any protein identified by SwissProt curators as 
having a GPI-anchor has been mapped to 'GPI-anchor binding'.  Apart from 
CD24a and Thy1, all the other genes in MGI annotated to 'GPI anchor 
binding' are based on this keyword mapping.  Looking at Amigo, I believe 
that every single experimentally based annotation to 'GPI anchor 
binding' is for a GPI anchored protein, so clearly other annotators use 
this term in this way.  I would bet this term was put in the GO by 
someone who intended it to be used this way.  And, as Karen so 
conveniently pointed out, the definitions of GO binding terms, including 
GPI anchor binding, do not specifically exclude covalent binding, so 
logically the annotations stand.  I would also point out that neither 
the 'Annotation Guide' nor 'Molecular Function Ontology Guidelines' 
sections of the documentation on the Gene Ontology Consortium's official 
web pages exclude annotation of covalent binding or even mention it.  We 
don't even exclude annotation of covalent binding in the much stricter 
MGI GO annotation guidelines (the written ones).

However, since I know that you are now at least driven halfway up the 
wall by my argumentation, I am may be willing to compromise and propose 
that in addition to eliminating the SwissProt keyword mapping, that we 
obsolete the term 'GPI-anchor binding ; GO:0048503' itself.  I know of 
no protein that binds GPI anchors specifically in a non-covalent way, 
although I agree it is a theoretical possibility, and the term clearly 
has only been a source of confusion to GO annotators, although the 
annotations are actually _useful_ for GO users, who might want to be 
able to easily find GPI-anchored proteins all at once.  But forget about 
the users -- with luck they may figure out that 'anchored to external 
side of plasma membrane ; GO:0031362' can be used to find some GPI 
anchored proteins, although there is no SwissProt mapping to this term 
and this term is less specific than 'GPI anchor binding'.

Rest assured Harold, I do not routinely annotate covalent binding, 
except in the case of 'GPI anchor binding' where I felt I was following 
the precedent of others in the GO and providing quite useful information 
to our users.  Although I am open to doing so, I do not plan remove my 
experimentally correct annotations until further discussions of this 
issue have taken place.

Thanks,

Alex


Harold Drabkin wrote:
> Alex can you please look at the GPI anchor binding annotations that 
> you made here. The term should only be used for non-covalent binding, 
> not covalent attachment. Since you recently looked at the papers, you 
> are more familiar with what they were trying to show.
>
> hjd
>
>
>
> -------- Original Message --------
> Return-Path:     <val at sanger.ac.uk>
> Received:     from smtp.sanger.ac.uk (smtp.sanger.ac.uk 
> [193.62.203.215]) by mirkwood.informatics.jax.org (8.13.8/8.13.8) with 
> ESMTP id m0GIBhtp026890 for <hjd at informatics.jax.org>; Wed, 16 Jan 
> 2008 13:11:43 -0500 (EST) (envelope-from val at sanger.ac.uk)
> Received:     from intmail1b.internal.sanger.ac.uk ([172.17.56.130] 
> helo=smtp.sanger.ac.uk) by routemail.internal.sanger.ac.uk with esmtp 
> (Exim 4.66) (envelope-from <val at sanger.ac.uk>) id 1JFCjl-0002W8-LF for 
> hjd at informatics.jax.org; Wed, 16 Jan 2008 18:11:41 +0000
> Received:     from deskpro16317.dynamic.sanger.ac.uk ([172.20.8.72]) 
> by intmail1b.internal.sanger.ac.uk with esmtp (Exim 4.41) id 
> 1JFCjl-0003rb-Fm; Wed, 16 Jan 2008 18:11:41 +0000
> Received:     from localhost.sanger.ac.uk ([127.0.0.1]) by 
> deskpro16317.dynamic.sanger.ac.uk with esmtp (Exim 4.50) id 
> 1JFCjl-0008GD-Ef; Wed, 16 Jan 2008 18:11:41 +0000
> Message-ID:     <478E48DD.1010503 at sanger.ac.uk>
> Date:     Wed, 16 Jan 2008 18:11:41 +0000
> From:     Valerie Wood <val at sanger.ac.uk>
> User-Agent:     Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.7.8) 
> Gecko/20070113 Debian/1.7.8-1sarge10
> X-Accept-Language:     en
> MIME-Version:     1.0
> To:     hjd at informatics.jax.org
> CC:     Valerie Wood <val at sanger.ac.uk>
> Subject:     Re: [annotation] RNA trimethylguanosine cap binding query
> References:     <478E3631.2030208 at sanger.ac.uk> 
> <478E435E.30002 at informatics.jax.org>
> In-Reply-To:     <478E435E.30002 at informatics.jax.org>
> Content-Type:     text/plain; charset=ISO-8859-1; format=flowed
> Content-Transfer-Encoding:     7bit
> X-PMX-Version:     5.4.0.320885, Antispam-Engine: 2.5.2.313940, 
> Antispam-Data: 2008.1.16.95857
> X-PerlMx-Spam:     Gauge=IIIIIII, Probability=7%, Report='__CT 0, 
> __CTE 0, __CT_TEXT_PLAIN 0, __HAS_MSGID 0, __MIME_TEXT_ONLY 0, 
> __MIME_VERSION 0, __SANE_MSGID 0, __USER_AGENT 0'
> Content-Length:     2374
>
>
>
> *Hi Harold,
>
> mouse
> Cd24a
> uses *PMID: 14707049 for what seems to be a covalent GPI-binding 
> doesn't it?
>
> i'm sure there is 'mixed usage'
>
> or is there some other data in there (I just did a search on GPI)
>
> Val
>
>
>
>
>
> Harold Drabkin wrote:
>
>> Valerie Wood wrote:
>>
>>> I recently used the term RNA trimethylguanosine cap binding  to 
>>> annotate pombe telomerase RNA and represent the fact that this is 
>>> trimethylguanosine capped, but on re-reading the definition  I'm not 
>>> sure if this is correct?
>>> Can I use this for the modification itself?  or is it for gene 
>>> products which interact with a capped product?
>>
>> It represents some gene product that binds to the triMeG cap present 
>> on an mRNA. Not the gene product that makes the cap. A binding term 
>> should never be used for a covalent bond. Binding implies a 
>> reversible reaction at STP, with a Ka and Kd.
>>
>> GPI anchor binding it meant to be used for something that binds the 
>> the GPI anchor and not covalently linked to it.
>>
>>>
>>> There do not appear to be any other annotations to this term 
>>> despite  the fact that many RNAs are capped which is another reason 
>>> which made me suspect my usage may be wrong.
>>>
>>> Should  the binding terms should only be used for non-covalent 
>>> modifications (although this is only in some of the binding defs?), 
>>> and  does not represent the use of some terms. For instance GPI 
>>> anchor binding is used for a number of proteins which are GPI 
>>> anchored, in addition to proteins which bins the GPI moiety during 
>>> GPI anchor biosynthesis.
>>>
>>> Cheers
>>>
>>> Val
>>>
>>> Def:
>>> Interacting selectively with the trimethylguanosine 
>>> (m(3)(2,2,7)-GTP) moiety located at the 5' end of some RNA 
>>> molecules. Such trimethylated cap structures, generally produced by 
>>> posttranscriptional modification of a 7-methylguanosine cap, are 
>>> often found on snRNAs and snoRNAs transcribed by RNA polymerase II, 
>>> but have also be found on snRNAs transcribed by RNA polymerase III. 
>>> They have also been found on a subset of the mRNA population in some 
>>> species, e.g. C. elegans.
>>>
>>>
>>
>>
>>
>>
>
>
>


-- 
Alexander Diehl, Ph.D.
Senior Scientific Curator
Mouse Genome Informatics
The Jackson Laboratory
600 Main Street
Bar Harbor, ME  04609

email:  adiehl at informatics.jax.org
work:  +1 (207) 288-6427
fax:  +1 (207) 288-6131




More information about the Annotation mailing list