Search Mailing List Archives


Limit search to: Subject & Body Subject Author
Sort by: Reverse Sort
Limit to: All This Week Last Week This Month Last Month
Select Date Range     through    

[Software-group] Programmatic access to GO data

Tanya Berardini tberardi at acoma.stanford.edu
Tue Jul 27 10:44:51 PDT 2010


Hello software folks,

I'm forwarding an email that came in through GO help.  Sounds like this
fellow was corresponding with Midori and Seth back in March just before
Midori left.  Can you pick up the conversation please?

Thanks very much,

Tanya

Tanya Berardini
TAIR Curator
www.arabidopsis.org


---------- Forwarded message ----------
From: <gohelp-bounces at genome.stanford.edu>
Date: Mon, Jul 26, 2010 at 11:11 AM
Subject: Forward of moderated message
To: tberardini at arabidopsis.org




---------- Forwarded message ----------
From: Luqman Hodgkinson <luqman at berkeley.edu>
To: GO Help List <gohelp at genome.stanford.edu>
Date: Sat, 24 Jul 2010 08:14:46 -0700
Subject: Fwd: GO term enrichment




Dear GO Help,

A partial solution with a web service had been given to me in March. Either
this is no longer functioning or I have forgotten how to use it. Moreover my
question about how to supply the background set given in my email to Dr.
Midori Harris below was never answered. Ideally there would be a web service
such as that supplied to me earlier that would accept a background set and a
gene product set, all using UniProtKB identifiers, and would access current
annotation and ontology files to return a list of enriched terms. Downloaded
files become stale and do not fit nicely into the programmatic framework I
am developing. This would be a great help!

Sincerely, with best wishes,
Luqman Hodgkinson,
Ph.D. student, UC-Berkeley



Begin forwarded message:

*From: *Luqman Hodgkinson <luqman at berkeley.edu>
*Date: *July 24, 2010 7:59:09 AM PDT
*To: *GO Help List <gohelp at genome.stanford.edu>
*Subject: **Fwd: GO term enrichment*





Dear GO Help,

After working on several other projects, I am now returning to the project
of computing GO enrichment programmatically. I would like to supply a
background set and a gene product set using UniProtKB identifiers, and, in
response, I would like to retrieve GO terms that are enriched in the gene
product set relative to the background set. Ideally, no downloads would be
required (e.g. no .obo files), but, rather, the most current version of GO
annotations would be used. Is this possible? It would be very useful for my
project.

Sincerely, with best wishes,
Luqman Hodgkinson,
Ph.D. student, UC-Berkeley




Begin forwarded message:

*From: *Luqman Hodgkinson <luqman at berkeley.edu>
*Date: *March 10, 2010 9:55:05 PM PST
*To: *Midori Harris <midori at ebi.ac.uk>
*Subject: **Re: GO term enrichment*





Dear Midori,

Thank you! Seth Carbon sent me the URL for the script which works very well.
I have one additional question. The URL is

http://ash.lbl.gov/cgi-bin/amigo/term_enrichment?output=normal&format=tab&request=results&force=yes&speciesdb=UniProtKB/Swiss-Prot&gp_list=

with the gene product list appended.

>From the base URL: http://ash.lbl.gov/cgi-bin/amigo/term_enrichment, it is
clear that both a gene product list and a background set can be supplied.
However, I do not know how to specify the background set in the URL so that
I can perform the enrichment analysis programmatically. There was no example
of this in the script.

 With best wishes,
 Luqman




On Mar 9, 2010, at 9:02 AM, Midori Harris wrote:

Dear Luqman,


I'll forward your question; sorry for the bother.


Midori


On Tue, 9 Mar 2010, Luqman Hodgkinson wrote:






Dear Midori,


Thank you so much for your helpful replies!! Just recently I tried to use
this suggestion which you sent to me from one of the programmers.
Unfortunately, the folder go-dev/amigo/scripts does not contain the script
that was mentioned: term_enrichment_batch.pl.


The contents of go-dev/amigo/scripts are


acotillo.pl             make_misc_key.pl

indexer.pl              make_spec_key.pl


none of which are the referred script.


May you be able to request the script from the original programmer who
referred it to you? It would be a great help to have this script!!


 With best wishes,

 Luqman





On Feb 4, 2010, at 2:32 AM, Midori Harris wrote:


Dear Luqman,


I've had this suggestion from one of our programmers:


--------

There is a perl script in the CVS go-dev repository that can perform batch
queries against the experimental server (don't want to hammer the main
site). It doesn't have the best stability, but it should be a great help in
getting through large numbers of queries. CVS instructions are at:


http://sourceforge.net/projects/geneontology/develop


The script is at:


go-dev/amigo/scripts/term_enrichment_batch.pl


I believe the only requirement is "LWP::UserAgent".


Cheers,


-Seth

--------


>From your original question I'm guessing that you're already aware that you
can install GO::TermFinder locally and use the supplied command line
scripts.


Best regards,

Midori


On Wed, 3 Feb 2010, Luqman Hodgkinson wrote:






Dear Midori,


Thank you for your speedy reply! I had seen the AmiGO browser and it seems
very useful. However I must perform this process 300 times. Is there a way
of automating the AmiGO method? Or is there another method you would
recommend in order to automate the process of performing this many GO
enrichment requests?


 With best wishes,

 Luqman





On Feb 3, 2010, at 8:19 AM, Midori Harris wrote:


Dear Luqman,

The Term Enrichment tool that is available on the web with the AmiGO browser
allows you to specify an input set of genes of interest simply by
typing/pasting them into the query form, or by uploading them in a file that
contains a simple list (using the gene_association file format is optional,
not required).

http://amigo.geneontology.org/cgi-bin/amigo/term_enrichment

Although we recommend using your own background gene set, it is also
possible to use the  database filter selection to choose a set from the
annotation sets in the GO repository. If you do upload your own background
set, it can be in the form of a simple list.

The web-based Term Enrichment tool does not require you to upload a .obo
file; it uses GO ontology data from the AmiGO database.

I hope this addresses your question, but please let us know if you have any
others.

Regards,

Midori

============================

Midori A. Harris, Ph.D.

GO Editor

EMBL - EBI

Wellcome Trust Genome Campus

Hinxton

Cambridge CB10 1SD UK

Tel: +44 (0) 1223 494667

Fax: +44 (0) 1223 494468

Email: midori at ebi.ac.uk

Date: Wed, 3 Feb 2010 07:59:08 -0800

Email: luqman at berkeley.edu Name: Software (from Luqman Hodgkinson)

Text: Dear Sir or Madam,

What is the best way for me to find enrichment of GO terms in a small set of
genes represented by UniProtKB IDs? I also have a large set of genes that
represent the entire universe over which I am searching. I wanted to use
GO:TermFinder, but it requires both an annotation file an a .obo file. How
may I obtain these files: annotation and .obo for my large set of genes?

With best wishes,

Luqman
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.stanford.edu/pipermail/go-software/attachments/20100727/24bbdfd7/attachment.html>


More information about the go-software mailing list