[Annotation] [Fwd: HIGH level GO mapping]

[David Lonsdale]

Hi all,
that seems clear, thanks for the comments. It would be useful to had a 
flag somewhere saying that the InterPro entry has been looked at, by 
whom and a timestamp, specifically where it cant be mapped. We 
(curators) have discussed this on many occasions but nothing has been 
done yet; it would be extremely useful.
David

edimmer at ebi.ac.uk wrote:
> Hi David, Val and Jennifer
>
> Yes, I agree with Val.  If no more granualar term can be located than the
> root terms, then am InterPro domain should not be mapped.
>
> My interpretation of the annotation to the root terms is a little
> different - which ever method is used - manual or electronic - all
> annotations to root terms simply state that no detail of its MF, BP or CC
> can be provided, and if the manual ND code is used, this indicates that no
> data can be found by a curator, whereas if the InterPro2GO reference is
> inserted, it means that no information can be provided using this method.
> In any case! Our two interpretations come to the same conclusion, that
> InterPro2GO mappings to such terms are unuseful to the user of the
> annotation sets. They may be handy for curators of the mapping, but I
> think we could look to having such information stored elsewhere :-)
>
> Emily
>
>   
>> Hi Emily,
>>
>> I suggested to David that these should probably not be mapped to
>> molecular function root node (I think there are some IEA root terms
>> mappings already, but I filter and discard them for pombe, on the basis
>> that all root node annotation should be manually curated). For most of
>> these IEA root node annotations, even though there is not a general
>> mapping which can be added  more specific annotations can be/are made,
>> so the root node annotations are confusing.
>>
>> In general, a root node annotation means that there is no information on
>> the F,P,C for the gene in question (albeit with ND). What the IEA
>> mapping to the root node is saying is that no generic IEA mapping can be
>> identified, which is slightly different. I would prefer to see no
>> mapping in these cases, reserving root node annotations for cases where
>> there really is no data, but I don't know if there is a general
>> consensus for this.
>>
>> I raised this as a divergent issue on this list, thread
>> Re: [Annotation] check for ND + additional annotation
>>  09/05/08 10:43
>>
>> Val
>>
>>
>> David Lonsdale wrote:
>>     
>>> Hi Emily,
>>>
>>> What is the policy on high level mapping in the absnce of other
>>> available terms
>>> e.g.
>>> GO:0008150 - Biological process
>>> GO:0003674 - Molecular function
>>>
>>> The case of entry IPR000994 comes to mind. It represents the Peptidase
>>> M24 fold. It was (is) named the 'Peptidase M24, catalytic core' ---- BUT
>>> and was mapped to peptidase activity.
>>>
>>> However, it has three basic functions:
>>> 1. protein binding
>>> 2. metal ion binding
>>> 3. hydrolase activity
>>>
>>> The entry contains :
>>> peptidases (1,2,3 - categories)
>>> creatinase  (1,3 - categories)
>>> Spt3 (Fact complex component) - ( 1 -category)
>>>
>>> Therefore it cannot be GO mapped to peptidase activity nor can it be
>>> GO mapped to 'hydrolase activity' nor 'catalytic activity'; but above
>>> that is 'molecualr function'. Can we map at that top level is the query.
>>>
>>> This is quite important as it will affect many entries that have an
>>> SSF identifying a structural fold that is involved in some form of
>>> catalytic activity.
>>>
>>> Look forward to your reponse.
>>>
>>> cheers David
>>>       
>> --
>> ---------------------------------------------------------------------------
>> Valerie Wood			 Tel: 01223 496909
>> S. pombe Genome Project		 Fax: 01223 494919
>> Wellcome Trust Sanger Institute	 email: val at sanger.ac.uk
>> Wellcome Trust Genome Campus	 http://www.genedb.org/genedb/pombe
>> Hinxton, Cambridge, CB10 1HH	 http://www.sanger.ac.uk/Projects/S_pombe
>>
>>
>>
>> --
>>  The Wellcome Trust Sanger Institute is operated by Genome Research
>>  Limited, a charity registered in England with number 1021457 and a
>>  company registered in England with number 2742969, whose registered
>>  office is 215 Euston Road, London, NW1 2BE.
>>
>>     
>
>   


-- 
'To err is human to really louse things up takes a computer'

David Lonsdale

Tel: +44 (0)1223 494694
Mob: 07974920437

Email: davidl at ebi.ac.uk