Release Name: 2.2.0
Notes:
Release Notes
=============
#Product:# caArray
#Version:# 2.2.0
#Date:# January 2009
Contents
--------
1. Introduction
2. Release History
3. Anticipated Releases
4. Features and Defects Addressed in this Release
5. Known Issues/Defects
6. Bug Reports, Feature Requests and Support
7. Documentation and Files
8. NCICB Web Pages
Introduction
---------------------------
caArray is an open-source, web and programmatically
accessible array data management system. caArray guides the
annotation and exchange of array data using a federated
model of local installations whose results are shareable
across the cancer Biomedical Informatics Grid (caBIG™).
caArray furthers translational cancer research through
acquisition, dissemination and aggregation of semantically
interoperable array data to support subsequent analysis by
tools and services on and off the Grid. As array technology
advances and matures, caArray will extend its logical
library of assay management.
* https://array.nci.nih.gov/
Release History
------------------------
* caArray v2.2.0 -- January 2009
* caArray v2.1.1 -- October 2008
* caArray v2.1.0 -- August 2008
* caArray v2.0.2 -- May 2008
* caArray v2.0.1 -- April 2008
* caArray v2.0.0 -- February 2008
Anticipated Releases
------------------------
* caArray v2.3.0 -- May 2009 (new features)
Features and Defects Addressed in this Release
-----------------------------------------------
This release represents a feature release of the caArray 2 software.
Approximately 80 features/issues were addressed in this release.
Highlights of caArray 2.2.0 are:
* Sample search:
* The user can now search for biomaterials (Samples and Sources)
in the system.
* The keyword search can be applied to one of a list of standard
or predefined categories, or to arbitrary categories previously
imported into the system via Characteristic[] columns in a
MAGE-TAB SDRF document.
* Export experiment annotations in MAGE-TAB format:
* The user can generate and download a MAGE-TAB file set describing
the annotations of an experiment.
* The generated MAGE-TAB includes the relationship between biomaterials,
hybridizations and data files, and also characteristics of the
biomaterials.
* The generated MAGE-TAB does not contain information about
experimental factors, protocols, publications or people.
* Bulk-update of experiment annotations using MAGE-TAB import:
* When a MAGE-TAB file set is being imported, the System recognizes
references to existing biomaterials and hybridizations by name.
* Biomaterial linkages can be modified in an additive way via
MAGE-TAB import. Existing linkages and biomaterials cannot be
deleted via MAGE-TAB import.
* Biomaterial characteristics can be modified or new characteristics
added via MAGE-TAB import.
* However, experimental factors and protocols cannot be added/changed
for existing biomaterials and hybridizations via MAGE-TAB import.
* As part of the upgrade to 2.2.0, existing data will be scrubbed
to make biomaterial and hybridization names unique within an experiment.
In other words, all Sources with the same name (within an experiment) will
be merged into the same Source. A similar procedure will be followed
for Samples, Extracts and Labeled Extracts. But multiple Hybridizations
with the same name (within an experiment) will be renamed to have
distinct names. (The installer allows the user to override this
default strategy and merge duplicate Hybridizations instead.)
* Support for GEO and ScanArray data:
* Data files can be designated as GEO SOFT or GEO GSM and then
imported (without parsing) and associated with samples.
* An array provider called ScanArray is now supported, and ScanArray CSV
data files can be imported (without parsing) and associated with samples.
* Enhanced usability of permissions management workflow:
* While setting sample-selective access for an experiment, the user can
do a keyword search for Samples against a set of categories, select
the Samples of interest, and grant access to the selected Samples as
a whole.
* Enhancements to the GUI installer to make more properties configurable.
* Ivy Dependency Management Support:
* caArray uses the Ivy tool for dependency management. This means that,
rather than explicitly checking in all the third party libraries caArray
depends on, it expresses those dependencies using a declarative XML syntax.
The ivy tool then takes care of resolving those dependencies and retrieving
them from a repository. NCI maintains an Ivy repository with approved
versions of various tools, and caarray is configured to pull from this
repository. Retrieved dependencies are cached on each developer's machine,
so that for a given build, only changed dependencies neeed to be retrieved.
As a result, checkouts are significantly more compact. This process is
integrated with the ant-based build, and transparent to the developer.
The one exception is the grid service, which still uses a locally
checked-in set of libraries. This is due to the fact that the caGrid
authoring tools, namely Introduce, are not yet configured to produce
ivy-enabled service definitions.
* API client jars:
* caArray 2.2.0 is backwards-compatible with programmatic API clients
using 2.1.0 jars. Please note that the API client libraries published
in this release are the same libraries that were published with
caArray 2.1.0. (caarray_client_2_1_0.zip and caarray_api_javadoc_2_1_0.zip)
Known Issues/Defects
------------------------
See the GForge tracker for existing open defects, community requests,
resolutions and feature requests. The following issues are
highlighted.
https://gforge.nci.nih.gov/tracker/?group_id=305
* Data that was imported before caArray 2.1.0 has an issue where derived
data files are not associated with the corresponding raw data files.
A SQL script is provided to scrub pre-2.1.0 data to correct raw-derived
data relationships. Please download the fix_raw_derived_relationships.zip
package from the caArray2 Gforge site. It contains a README.txt and a
SQL script. It is highly recommended that you run this SQL script against
your database, since otherwise, the MAGE-TAB export of some of your
pre-2.1.0 experiments can be incorrect.
* The total import job size (defined as sum of uncompressed file sizes)
must be less than 3GB.
* A collaborator cannot see files in the "Uploaded" state even with
read-write permissions to the experiment.
* The installer does not support configuring SSL support in JBoss.
* After installing a local copy of caArray, you must go to
http://<IP_address_or_host_name>:<port>/caarray in order to test
your installation. Even if you are running the browser on the same
machine as your caArray installation, you cannot go to
http://localhost:<port>/caarray.
* Image files referenced in a MAGE-TAB SDRF cannot be validated or
imported.
* Multiple MAGE-TAB imports can result in duplicate persons in the database.
Bug Reports, Feature Requests, And Support
------------------------------------------
Send email to ncicb@pop.nci.nih.gov to request technical support. To
report a bug or request a new feature, please visit the Molecular Analysis Tools
Knowledge Center resources at:
https://cabig-kc.nci.nih.gov/Molecular/KC/index.php/Main_Page
Existing requests and resolution may be viewed at the caArray GForge URL:
https://gforge.nci.nih.gov/tracker/?group_id=305
Documentation And Files
-----------------------
Links to all documentation and files can be found at:
http://caarray.nci.nih.gov/
NCICB Web Pages
---------------
* The NCI Center for Bioinformatics, http://ncicb.nci.nih.gov/
* NCICB Application Support, http://ncicb.nci.nih.gov/NCICB/support
* NCICB Download Center, http://ncicb.nci.nih.gov/download/
Changes:
|
