[BiO BB] New release of STACKdb, in relational database format

[Liza Groenewald]

Major revamp of the STACKdb Human Gene Expression Index, provided free
to academics, incorporates mRNA sequences and visualization software.
----------------------------------------

The latest version of the STACK database of reconstructed human
expressed transcripts, STACKdb v3.0, has been released by The South
African National Bioinformatics Institute (SANBI), in collaboration with
Electric Genetics. The STACK human gene expression index is unique in
that it is the only gene expression index that is availabe with the
tools that are used in its manufacture. In addition, it contains
verified, clone-linked cluster assemblies. The index is processed into
hierarchically organized human transcript consensus sequences that
reflect gene expression forms and, in particular, alternate expression
variants, within several tissue-based and one disease expression
category.

STACKdb v3.0 is created from 2,524,209 EST sequences and 51,638
partially redundant mRNA sequences, distributed according to tissue
expression, and represents an increase of 569,014 new EST sequences and
51,638 mRNA sequences as compared to the previous release, STACKdb v2.4.
STACKdb v3.0 contains 241,258 clusters and 12,943 clonelinks in total.


What's New
 ---------------------------------------
STACKdb 3.0 has been produced with the latest generation stackPACK
transcript reconstruction and variation analysis management tools, and
therefore has several fundamental improvements:

* Inclusion of mRNA sequences in clusters.
* The database is now distributed in two forms:
    -  Relational database tables, for use with the web-based viewing
software.
    -  FastA files containing non-redundant sets of linked clusters,
clusters and singletons for each tissue type, for use with external
programs.
* Web-based viewing software, equivalent to that found in stackPACK
v2.1.1, which:
    -  highlights variation.
    -  allows the viewing of cluster, linked cluster and alternate
consensus sequences.
    -  displays initial (PHRAP) alignments, alignment analyses (CRAW
output), as well as alignments for each potential alternate expression
form.
* Various web-based and command line reports to extract data for use in
other programs.


Availability
-------------------------------
The database is available free of charge to academic and non-profit
institutions.

Interested users who wish to download STACKdb 3.0 may register at
www.sanbi.ac.za/CODES.  The Swiss Institute of Bioinformatics
collaborates with SANBI and serves SANBI's STACKdb download from the
Swiss Institute of Bioinformatics web site to ensure faster download
times.

The database may also be searched on-line by comparing a query sequence
to any of the STACKdb categories. WebBLAST is available at:
http://juju.egenetics.com/stackpack/webblast.html

The stackPACK tools, used to create STACKdb, are also available free of
charge to academics.  Interested users may register at
www.sanbi.ac.za/CODES to download the stackPACK tools for COMPAQ, Linux,

Silicon Graphics or Sun platforms.


Interested in transcript reconstruction?
-----------------------------------------------
The STACKers mailing list is for information on EST clustering and
transcript reconstruction. Please point your browser to the following
URL in order to subscribe to STACKers:
http://fling.sanbi.ac.za/mailman/listinfo/stackers


Scientific Publications
-------------------------------------------
StackPACK has been described in several publications.

* The method employed and assessment of quality of the resource is fully
described in:
STACK: sequence tag alignment and consensus knowledgebase.
Christoffels A, Gelder Av, Greyling G, Miller R, Hide T, Hide W.
Nucleic Acids Res. 2001 Jan 1;29(1):234-8.

and

A Comprehensive Approach to Clustering of Expressed Human Gene Sequence:

The Sequence Tag Alignment and Consensus Knowledge Base
Robert T. Miller, Alan G. Christoffels, Chella Gopalakrishnan, John
Burke, Andrey A. Ptitsyn, Tania R. Broveak, and Winston A. Hide.
November 1999: Genome Research: 9:1143-1155

* The clustering algorithm employed in stackPACK is described in:
d2_cluster: A Validated Method for Clustering EST and Full-Length cDNA
Sequences.
John Burke, Dan Davison, and Winston Hide.
November 1999: Genome Research: 9:1135-1142

Additionally, stackPACK has been instrumental in several discoveries. A
listing of scientific publications and posters involving stackPACK can
be found at:
http://www.egenetics.com/?Section=publications&Parent=science


Contact Details
----------------------------
STACKdb is managed and distributed by Electric Genetics who also
provides free support for academic users. Please contact
support at egenetics.com for support, feedback or any queries.

For further information on STACKdb, please refer to the Electric
Genetics website:
http://www.egenetics.com/?Section=STACKdb&Parent=products

For further information on research at the South African National
Bioinformatics Institute, please visit:
http://www.sanbi.ac.za/Research.html