Hi George
Unfortunately it is not possible to select the canonical transcript
from the Ensembl marts at present. It is possible to do this using the
Ensembl Perl API and if you send a ticket to [email protected] we
will be able to help you retrieve the information you need.
Regards
Rhoda
On 4 May 2012, at 00:46, George Gutman wrote:
Arek,
Thanks, that makes sense, I had imagined that the choice of "Unique
results" would have avoided this duplication. But this creates a
problem
for me, since there will be a large amount of redundancy in the
sequences
I recover, and I require a collection of unique sequences even at
the cost
of the database being less than complete. Is there a way for me to
avoid
this redundancy by requesting only a single transcript for each gene,
perhaps the longest or the most abundant?
I am requesting the email help you suggest.
Thanks,
George
On Thu, 3 May 2012, Arek Kasprzyk wrote:
Hi George,
the count " 21405/56478" refers to number of genes that have
protein coding
transcripts. A large proportion of these genes will have more than
one
transcript thanks to the alternative splicing model that Ensembl
supports.
Hence you will be getting far more transcripts than genes. If you
need to
know more about how Ensembl predicts genes and corresponding
transcripts
please drop an email to [email protected]
a
On Thu, May 3, 2012 at 1:59 PM, George Gutman <[email protected]>
wrote:
I'm trying to collect all protein coding sequences from various
species.
I start here: http://www.biomart.org/biomart/martview/
My selections are as follows:
Database: Ensembl Genes 66 (Sanger UK)
Dataset: Homo sapiens genes (GRCh37.p6)
Filters: Gene Type: protein coding
Attributes: Sequences/Coding Sequences
Header: Associated Gene Name, Description, Ensembl Gene ID
When I click on "Count" I get 21405/56478, which I interpret as
21405
coding sequences that I should be recovering out of a total of 56478
entries.
I click "Results", "Unique Results Only", "Export" to "Compressed
Web file
(notify by email), I enter my email address, then "Go".
The resulting file I download is 111,827 Kb in size and it
contains 98,024
entries, many more than the 21,405 I expected. (I determined this
by
doing a "seach and replace" for ">"). The first entry is gene name
"CYP26B1", and I find a total of five separate entries with this
name and
with identical gene descriptions and Ensemble Gene IDs. The five
sequences, however, are different, although three of them are
identical
through the first few lines.
So what am I doing wrong? When I went through the same steps for
E. coli
K12 I recovered a file with 4258 sequences, the number I expected
based
on the output of "Count".
Thanks,
George Gutman
Cheers,
George Gutman
*************************************************************************
* George A. Gutman, Professor
Emeritus *
* Department of Microbiology and Molecular
Genetics *
* University of California,
Irvine *
* Office: (949)824-6593 B250, Med
Sci *
* (714)552-1242 (cell) e-mail:
[email protected] *
* http://www.ucihs.uci.edu/microbio/facultyResearch/faculty/gutman.html
*
*************************************************************************
On Thu, 3 May 2012, Arek Kasprzyk wrote:
Date: Thu, 3 May 2012 12:55:22 -0400
From: Arek Kasprzyk <[email protected]>
To: George Gutman <[email protected]>
Cc: [email protected]
Subject: Re: [BioMart Users] Where to address questions
Dear George,
please feel free to post here and we'll try to help you. Please
give us
more details on your problem, the portal that you are using etc
a
On Wed, May 2, 2012 at 2:30 AM, George Gutman <[email protected]>
wrote:
I'm trying to download collections of protein coding regions from
various
genomes and am having trouble. Where can I address my
questions? Do I
need to be a member of this list to post here?
George Gutman
_______________________________________________
Users mailing list
[email protected]
https://lists.biomart.org/mailman/listinfo/users
--
Arek Kasprzyk, MD, MSc, PhD
BioMart Project Lead
www.biomart.org
_______________________________________________
Users mailing list
[email protected]
https://lists.biomart.org/mailman/listinfo/users
Rhoda Kinsella Ph.D.
Ensembl Production Project Leader,
European Bioinformatics Institute (EMBL-EBI),
Wellcome Trust Genome Campus,
Hinxton
Cambridge CB10 1SD,
UK.
_______________________________________________
Users mailing list
[email protected]
https://lists.biomart.org/mailman/listinfo/users
