WormBase release WS164 now online

This is an automatic announcement that WormBase
(http://www.wormbase.org) has just been updated.  New releases occur
roughly every three weeks.

The text of the AceDB release notes, which contains highlights of the
new data is attached.  You can download the full AceDB files from:

   ftp://ftp.sanger.ac.uk/pub/wormbase/current_release/
   or
   ftp://ftp.wormbase.org/pub/wormbase/acedb/current_release

Additional information on the release, including any necessary patches or bug
fixes can be found on the WormBaseWiki:

   http://www.wormbase.org/wiki/index.php/WSWS164

New release of WormBase WS164, Wormpep164 and Wormrna164 Wed Sep  6 13:39:15 BST 2006


WS164 was built by Gary Williams
======================================================================

This directory includes:
i)   database.WS164.*.tar.gz    -   compressed data for new release
ii)  models.wrm.WS164           -   the latest database schema (also in above database files)
iii) CHROMOSOMES/subdir         -   contains 3 files (DNA, GFF & AGP per chromosome)
iv)  WS164-WS163.dbcomp         -   log file reporting difference from last release
v)   wormpep164.tar.gz          -   full Wormpep distribution corresponding to WS164
vi)   wormrna164.tar.gz          -   latest WormRNA release containing non-coding RNA's in the genome
vii)  confirmed_genes.WS164.gz   -   DNA sequences of all genes confirmed by EST &/or cDNA
viii) cDNA2orf.WS164.gz           -   Latest set of ORF connections to each cDNA (EST, OST, mRNA)
ix)   gene_interpolated_map_positions.WS164.gz    - Interpolated map positions for each coding/RNA gene
x)    clone_interpolated_map_positions.WS164.gz   - Interpolated map positions for each clone
xi)   best_blastp_hits.WS164.gz  - for each C. elegans WormPep protein, lists Best blastp match to
                            human, fly, yeast, C. briggsae, and SwissProt & TrEMBL proteins.
xii)  best_blastp_hits_brigprot.WS164.gz   - for each C. briggsae protein, lists Best blastp match to
                                     human, fly, yeast, C. elegans, and SwissProt & TrEMBL proteins.
xiii) geneIDs.WS164.gz   - list of all current gene identifiers with CGC & molecular names (when known)
xiv)  PCR_product2gene.WS164.gz   - Mappings between PCR products and overlapping Genes


Release notes on the web:
-------------------------
http://www.wormbase.org/wiki/index.php/Release_notes



Genome sequence composition:
----------------------------

       	WS164       	WS163      	change
----------------------------------------------
a    	32365888	32365888	  +0
c    	17779857	17779857	  +0
g    	17756012	17756012	  +0
t    	32365687	32365687	  +0
n    	0       	0       	  +0

Total	100267444	100267444	  +0


Chromosomal Changes:
--------------------
There are no changes to the chromosome sequences in this release.


Gene data set (Live C.elegans genes 23787)
------------------------------------------
Molecular_info              22066 (92.8%)
Concise_description          4255 (17.9%)
Reference                    6323 (26.6%)
CGC_approved Gene name       8851 (37.2%)
RNAi_result                 19804 (83.3%)
Microarray_results          19123 (80.4%)
SAGE_transcript             19738 (83%)




Wormpep data set:
----------------------------

There are 20073 CDS in autoace, 23180 when counting 3107 alternate splice forms.

The 23180 sequences contain 10,177,179 base pairs in total.

Modified entries               7
Deleted entries               13
New entries                   28
Reappeared entries             1

Net change  +16



Status of entries: Confidence level of prediction (based on the amount of transcript evidence)
-------------------------------------------------
Confirmed              7791 (33.6%)	Every base of every exon has transcription evidence (mRNA, EST etc.)
Partially_confirmed   10753 (46.4%)	Some, but not all exon bases are covered by transcript evidence
Predicted              4636 (20.0%)	No transcriptional evidence at all



Status of entries: Protein Accessions
-------------------------------------
UniProtKB/Swiss-Prot accessions   3270 (14.1%)
UniProtKB/TrEMBL accessions     19550 (84.3%)



Status of entries: Protein_ID's in EMBL
---------------------------------------
Protein_id            22820 (98.4%)



Gene <-> CDS,Transcript,Pseudogene connections (cgc-approved)
---------------------------------------------
Entries with CGC-approved Gene name   7185


GeneModel correction progress WS163 -> WS164
-----------------------------------------
Confirmed introns not in a CDS gene model;

		+---------+--------+
		| Introns | Change |
		+---------+--------+
Cambridge	|     19  |     2  |
St Louis 	|     10  |     0  |
		+---------+--------+


Members of known repeat families that overlap predicted exons;

		+---------+--------+
		| Repeats | Change |
		+---------+--------+
Cambridge	|      6  |     0  |
St Louis 	|      6  |     0  |
		+---------+--------+



Synchronisation with GenBank / EMBL:
------------------------------------

No synchronisation issues


There are no gaps remaining in the genome sequence
---------------
For more info mail [email protected]
-===================================================================================-



New Data:
---------

Mass spectrometry peptide data

 Mass spectrometry peptide data from Lukas Reiter at Michael
 Hengartner's laboratory in the University of Zürich has been added to
 this release.  Proteins from various fractions of C.elegans
 preparations and from various stages of the life cycle have been
 fragmented and the molecular weights of the fragments have then been
 used to deduce the sequence of the fragments by comparing to a
 database of C. elegans protein sequences using the package Sequest.

 The results were statistically verified with PeptideProphet and
 ProteinProphet (these are tools developed in the group of R.
 Aebersold who is working at the ETH in Zürich)

 The peptides were mapped to the genome using the known positions of
 the peptides in the proteins and the known positions of the proteins'
 genes on the genome.

 About 7,500 C. elegans genes have one or more mass spectrometry
 peptides mapped to them.

 This data gives confirmatory evidence of many exons and (where they
 span splice sites) some introns. It gives data on the stage of the
 life cycle and location in the organism that proteins are expressed.

     
COMPARA data

 New EnsEMBL COMPARA based predictions are included in WS164 and
 replace the older orthologue predictions. As the EnsEMBL COMPARA is
 now included in the regular Wormbase pipeline the predictions will be
 updated during each following build.

 The COMPARA orthologue prediction algorithm uses a combination of
 bidirectional best hits and conserved gene order to determine
 orthologueous genes and is available as part of the EnsEMBL codebase.

 A dump (mySQL) of the EnsEMBL databases used for the predictions is
 available from the Sanger ftp-site.

 data file: <ftp.sanger.ac.uk/pub2/wormbase/data/compara_164.tar.bz2>
 check-sum: <ftp.sanger.ac.uk/pub2/wormbase/data/compara_164.md5>

 The homologous genes represent the best reciprocal BLAST hits for the
 two species with additional pairs obtained by a combination of BLAST
 and location information for more closely related species. These
 homologues may therefore potentially represent orthologues. The
 Compara gene orthology predictions pipeline works at the protein
 sequence level and involves:

 We only analyze protein-coding genes, skipping all pseudogene
 predictions and non-coding RNA gene types. Due to alternate splicing
 of exons, one gene can produce multiple transcripts and hence
 multiple translations. In order to provide orthology at the gene
 level, one protein sequence must be assigned to the gene. The Ensembl
 compara pipeline picks the longest translation. This set of longest
 gene translations is now the proteome for each genome, which is
 analyzed.

 For each genome, the set of longest gene translations are dumped into
 FASTA files to be used as blastp databases. We use the Washington
 University version 2.0 of BLASTP with the following command line
 options: -filter none -span1 -postsw -V=20 -B=20 -sort_by_highscore -
 warnings -cpus 1. The -postsw option will perform full Smith-Waterman
 alignment of sequences and re-rank the database matches accordingly
 prior to output. We run BLASTP with only one query sequence per job
 resulting in a large number of BLASTP jobs managed by a processing
 system based on autonomous workers the Ensembl hive. The BLASTP
 output is then post-filtered so that only hits with an expectation
 (E)-value less than 1e-10 are stored in the compara database.

 The orthologue prediction algorithm is based on the concept of Best
 Reciprocal Hits. The analysis is done on a genome pair basis, for
 example a search for human - mouse orthologues.  Each gene's longest
 translation will likely hit the target genome in multiple locations.
 The 'best' hit is the one with the highest BLASTP score, followed by
 lowest E-value, followed by highest percent identity, followed by
 highest percent positivity. Since the 'best' is a simple sort, the
 score takes precedence over the other measures. Due to gene
 duplications, although rare, a query translation may align with
 identical score, E-value, percent identity, and percent positivity to
 more than one target translation. This can then result in 'ties' for
 the best position.

 For closely related species (i.e. Caenorhabditae), where some gene
 order conservation is expected, we identify additional orthologous
 pairs obtained by a combination of reciprocal BLAST and location
 information. This results in a reciprocal pair, where one direction
 is the best hit, but the reverse hit is less than best. To classify
 as orthologue the pair must also maintain synteny (conserved gene
 order) within 1.5 MB of another orthologue pair.


ncRNA data

 Included in this and subsequent WormBase releases is a set of 3672
 predicted ncRNA genes that are viewable through the genome browser
 <http://wormbase.sanger.ac.uk/db/seq/gbrowse/wormbase/>

 This set of genes was predicted using the RNAz programme
 <http://www.tbi.univie.ac.at/~wash/RNAz/> and published by K. Missal
 et al. J Exp Zoolog B Mol Dev Evol. 2006 Jul 15;306(4):379-92.
 <http://wormbase.sanger.ac.uk/db/misc/paper?name=WBPaper00027050;class=Paper>
 The objects are not stored in the ACeDB release of the database but
 this is a large set of predictions that overlaps with the WormBase
 curated ncRNA gene set.


Genome sequence updates:
-----------------------

None.


New Fixes:
----------

None.


Known Problems:
---------------

None.


Other Changes:
--------------

The phenotypic term "No_abnormality_scored," has been made obsolete
effective as of this release (WS164).  Instead, "Abnormal" will be
used with the "Not" qualifier instead of "No_abnormality_scored."
"Abnormal" will then be the root term for the entire phenotype
ontology.

Proposed Changes / Forthcoming Data:
-------------------------------------


Model Changes:
------------------------------------

    Changed ?Paper.Reference.Year to DateType rather than Int
    Existing data that is year as Int (eg 2006) will still be valid 

-===================================================================================-


Quick installation guide for UNIX/Linux systems
-----------------------------------------------

1. Create a new directory to contain your copy of WormBase,
	e.g. /users/yourname/wormbase

2. Unpack and untar all of the database.*.tar.gz files into
	this directory. You will need approximately 2-3 Gb of disk space.

3. Obtain and install a suitable acedb binary for your system
	(available from www.acedb.org).

4. Use the acedb 'xace' program to open your database, e.g.
	type 'xace /users/yourname/wormbase' at the command prompt.

5. See the acedb website for more information about acedb and
	using xace.

____________  END _____________