[BiO BB] API for gene ontology retrieval?

[Ahmed Moustafa]

Thank you all so much for your replies!!!
I think Chri's approach, to download the database and deal with it, is 
the way to go.
Do you know whether the GO database holds beyond the models organisms?
Or would I have to restrict the BLAST search to a model organism to be 
able to retrieve the gene ontology later from the GO database for the hit?
BTW, is there a plan to make the GO database remotely accessible (e.g. 
Ensembl core database)?
Once again, thank you so much!
Ahmed

On 11/10/2006 12:22 AM, Chris Mungall wrote:

>
> You can download the GO MySQL database and access it using the go-db- 
> perl API. This assumes you already have identifiers for your best  
> hits, and those identifiers have annotations associated with them. At  
> this time the API does no perform any kind of enrichment analysis; if  
> you want to do this sort of thing have a look at some of the tools on  
> the GO website.
>
> There is no web service programmatic access for the database right  
> now - you have to download it and access it that way. We'll hopefully  
> have some kind of web service available soon, perhaps a SPARQL endpoint.
>
> Details for the API can be found here:
> http://www.godatabase.org/dev
>
> On Nov 7, 2006, at 5:43 PM, Rob C wrote:
>
>> If performance is not a consideration, you may try irobotsoft from  
>> http://irobotsoft.com.
>> It has a visual interface for programming web robots, and also has  
>> an ActiveX interface
>> for programmers.
>>
>> They have an example 'blastx.irb' for the blast search automation.
>>
>> Yours,
>> rob
>> On 11/7/06, Ahmed Moustafa <ahmed at pobox.com> wrote: Hi All,
>> Is there some API or an automated method to retrieve the annotation
>> (gene ontology?) of a set of sequences?
>> I believe BLAST is necessary but after BLASTing the sequences, how  
>> can I
>> get the annotations (e.g. cellular component, pathway, biological
>> function...) of (for example) the best hit for each sequence?
>> Your help will be appreciated so much!
>> Thanks in advance,
>> Ahmed
>