gmod-1.1 (Chado) released

classic Classic list List threaded Threaded
1 message Options
Reply | Threaded
Open this post in threaded view

gmod-1.1 (Chado) released

Scott Cain

I am pleased to announce that we released gmod-1.1 today.  This
release has the GMOD database schema, Chado, and a variety of tools
for loading and working with the database.  The release can be
downloaded from the GMOD SourceForge site at

There have been many people who have helped with getting this release
done and I would like to thank all of them.  I would like to
especially thank Rob Buels and Naama Menda and the rest of the people
at the Sol Genomics Network (SGN, for their
code contributions, including a pub table loader that works with
PubMed and an updated version of the script to calculate cvterm

There have been many changes and bug fixes since the 1.0 release just
over two years ago, but the schema itself has changed very little.
For a SQL summary of the changes, see Changes.sql in the release that
contains SQL statements that should take a 1.0 release of the Chado
schema to the 1.1 schema.  If you have been using Chado from a recent
svn checkout, you don't need to modify your schema at all, since the
last change to the schema was about a year ago.

Now that this release is out, the plan going forward is for frequent,
smaller releases, with an eye on the Natural Diversity schema
modifications being worked on now.

Thanks for your continued support of the GMOD project,

* Fixes to so that file splitting will work
as expected.
* Minor fix to Bio::GMOD::DB::Adaptor (used by
to silence a uninint warning.
* Created a plpgsql function to share exons (called share_exons) that
takes exon features that have the same boundaries as another exon and reduces
them to one feature and fixes up feature_relationship entries so that
the remaining exon belongs to the right transcript/genes.  This may
be needed for Apollo.  Thanks to Robin Houston at GeneDB/Sanger Centre
for contributing much of the code.
* Created a function to order exons (called order_exons) that will create
entries in the feature_relationship.rank column to number exons (0 based)
from beginning to end in a given transcript.  This was originally needed
for Apollo, but by the time this function was done, Ed Lee had fixed up
Apollo to not need it.
* Fixed the audit module so that the triggers in it work with versions
of PostgreSQL greater that 8.0.
* Added so-bridge views.  These are a set of views on the feature table
the use SO terms for the names of the views, making it easy to query for
genes by doing something like this:  SELECT * FROM so.gene WHERE name='foo'.
See for more information.
* Added a protein_code_gene view (depends on so-bridge); thanks to Chris
* Updated all_feature_names view to search featureprop.value and
dbxref.accession and added organism_id so GBrowse can limit
by organism.
* Fixed to deal correctly with the case that
mulitple organisms have the same common name.  If the user tries to
use a common name that corresponds to more than one organism, it will
die gracefully.
* Fixed so that featureloc.srcfeature_id will
not be sometimes set incorrectly when there are multiple organisms
with chromosomes (or other srcfeatures) with the same name.
* Added a __DIE__ signal handler to so that the
databsae will be cleaned up in most cases before the loader croaks.
* Fixed so that when loading dna sequence from the
fasta section it won't accidentally add sequence to a feature that has
the same uniquename but different organism_id to the actual target.
* Adding a update_sequences method for use by
that will check the next value available to the database sequences
that generate primary keys and update them when necessary.  This
can happen when another database load adds data to chado but does
not update the sequence.
* Adding the ability to specify that use the
system tmp directory rather than using the current working directory
for writing the temporary load files.
* Added a --fastafile option to to use when
loading fasta files.
* Removed the cause of the error messages at the end of ontology loading:
it was a patch to correct for bad behavior that DBIx::DBStag which was
fixed some time ago.
* Added contributed by SGN.
* Added contributed by SGN.

Scott Cain, Ph. D.                                   scott at scottcain dot net
GMOD Coordinator (                     216-392-3087
Ontario Institute for Cancer Research


Gmod-schema mailing list
[hidden email]