[Bioperl-l] Clustering of Gene Ontology Terms

[Chris Fields]

On Feb 21, 2008, at 6:29 AM, K. Shameer wrote:

> Dear All,
>
> Thanks for your reply.
>
>> I'm not sure what you mean by GO clustering.  However, there are a
>> number of packages that will look for enriched GO terms given a set  
>> of
>> Gene IDs.
>
> In simple terms I have a list of genes, irrespective of any  
> experimental
> value I need to cluster them based on their GO-Terms using the Di  
> Acyclic
> Graph data structure of the Gene ontology databases.
>
> I found an interesting package from CPAN : GO::TermFinder .
> http://search.cpan.org/dist/GO-TermFinder/
>
> I am not sure if this is a part of Bioperl :). I have tried with few
> examples it is working fine but I got few wrong results. If this is  
> not
> part of Bioperl project, I may need to write to the authors.

It's not part of BioPerl, so you should contact the author.

>> If you are looking for programmatic support for custom
>> pipelines, you can look at R/Bioconductor.
>
> Thanks a lot !! I have seen a couple of R packages related to GO.  
> But, I
> honestly wanted to do it in Perl :) !!!
>>

BioPerl's ontology classes are all in Bio::Ontology and  
Bio::OntologyIO.  I would also recommend looking at Chris Mungall's go- 
perl:

http://search.cpan.org/~cmungall/go-perl-0.08/go-perl.pod

>> Here is a website that will look for GO enrichment and takes Flybase
>> IDs as the gene identifier:
>> http://niaid.abcc.ncifcrf.gov/home.jsp
>
> Thanks for the link. But I am specifically for something stand alone  
> due
> to huge volume of my dataset.
>
> Cheers,
> Shameer
> NCBS - TIFR

It's always faster to do it locally, I agree, as long as the tools are  
available to do so.

chris