Search Mailing List Archives
[go-helpdesk] [go-software] Multiple results for single human gene name
sjcarbon at lbl.gov
Tue Jun 19 11:11:31 PDT 2012
Although I think there are different levels of "in GO" and I'm not quite
sure what he's getting at, if he's just looking to see what identifiers
are in the database, queries like:
dbxref.xref_dbname = 'HGNC';
might be fine. Associated terms through sequences can be seen with
term.id = association.term_id AND
association.gene_product_id = gene_product.id AND
gene_product.id = gene_product_seq.gene_product_id AND
gene_product_seq.seq_id = seq.id AND
seq.id = seq_dbxref.seq_id AND
seq_dbxref.dbxref_id = dbxref.id AND
dbxref.xref_dbname = 'HGNC'
Although that's a good third of the ontology with direct
annotation--more with the closure.
On 06/15/2012 03:28 AM, Peter Stenson wrote:
> Dear Tanya
> Thank you for your reply.
> My goal is to be able to produce a list of ontology terms associated
> with a gene (not a protein or protein fragment).
> Since UniProt catalogues proteins not genes, how should I uniquely
> identify a gene in AmiGO without resorting to calculating protein
> lengths from UniProt, and using the longest one, or relying on the most
> extensive record in the database as being the one to use?
> I guess I am asking if there is a list of unique gene identifiers (e.g.
> HGNC numbers) in GO?
> Many thanks
> Best regards
> -----tberardi at arabidopsis.org wrote: -----
> To: stensonpd at cardiff.ac.uk
> From: Tanya Berardini
> Sent by: tberardi at arabidopsis.org
> Date: 06/13/2012 07:02PM
> Cc: gohelp <go-helpdesk at lists.stanford.edu>, software-group
> <software-group at genome.stanford.edu>
> Subject: Multiple results for single human gene name
> Dear Peter,
> Thank you for contacting GO with your question. The six different human
> WHSC1 results in AmiGO correspond to six different UniProtKB records
> that are all called WHSC1. If you click on each name, you will see that
> the links to UniProtKB are all different.
> Four of them have this tag:
> "The sequence shown here is derived from an Ensembl automatic analysis
> pipeline and should be considered as preliminary data. "
> The remaining two differ in the length of the amino acid sequence.
> Depending on your goals, you may want to focus on just one or several of
> these records. This one, http://www.uniprot.org/uniprot/O96028, is the
> longest amino acid sequence and appears to be the most extensive
> record. If you use this UniProtKB identifier in your AmiGO search, you
> will only get one result.
> Best wishes,
> Tanya Berardini
> ---------- Forwarded message ----------
> From: stensonpd at cardiff.ac.uk <mailto:stensonpd at cardiff.ac.uk>
> To: go-helpdesk at mailman.stanford.edu
> <mailto:go-helpdesk at mailman.stanford.edu>
> Date: Wed, 13 Jun 2012 02:50:49 -0700
> Subject: GO Help query (from website)
> Email: stensonpd at cardiff.ac.uk <mailto:stensonpd at cardiff.ac.uk>
> Name: Ontology (from Peter Stenson)
> Text: Hello
> I am having trouble uniquely identifying genes in the GO database
> (online and download).
> For example, the search for human WHSC1
> brings up 6 results, all of which seem to be pointing to the same gene.
> Can you explain why there are 6 entries here, rather than just 1? Many
> go-software mailing list
> go-software at lists.stanford.edu
More information about the go-helpdesk