URL: http://probe.nalusda.gov:8000/acedocs/acedbfaq.html
Archive-name: acedb-faq
Last-modified: 30 Jun 98
Version: 1.42
------------------------------------------------------------------------
ACEDB FAQ
------------------------------------------------------------------------
Curated by: Dave Matthews
------------------------------------------------------------------------
Frequently Asked Questions about ACEDB
* Q0 : What is ACEDB?
* Q1 : ! What is the current version of ACEDB?
* Q2 : Where can I get ACEDB?
* Q3 : What hardware/software do I need to run ACEDB?
* Q4 : ! Can ACEDB be networked?
* Q5 : What documentation exists for ACEDB?
* Q6 : What newsgroups and mailing lists are available for ACEDB?
* Q7 : Is there a repository of software tools for ACEDB curators?
* Q8 : When and where is the next ACEDB Workshop?
* Q9 : How does ACEDB compare to commercial relational DBMS's?
* Q10 : How should ACEDB be cited?
* Q11 : What ACEDB databases exist?
* Q12 : Who prepared this document & where is the current version?
Questions marked with '+' are new, those with '!' have substantially changed
answers.
------------------------------------------------------------------------
Q0: What is ACEDB?
A0:
ACEDB is an acronym for A Caenorhabditis elegans Database. It can refer to a
database and data concerning the nematode C. elegans, or to the database
software alone. This document is concerned primarily with the latter
meaning. ACEDB is being adapted by many groups to organize molecular biology
data about the genomes of diverse species.
ACEDB allows for automatic cross-referencing of items during loading and
allows for hypertextual navigation of the links using a graphical user
interface and mouse. Certain special purpose graphical displays have been
integrated into the software. These reflect the needs of molecular
biologists in constructing genetic and physical maps of genomes.
ACEDB was written and developed by Richard Durbin (MRC LMB Cambridge,
England) and Jean Thierry-Mieg (CNRS, Montpellier, France), beginning in
1989. It is written in the C programming language and uses the X11 windowing
system to provide a platform independent graphical user interface. The
source code is publicly available. Durbin & Thierry-Mieg continue to develop
the system, with contributions from other groups.
A description by Durbin & Thierry-Mieg: ACEDB does not use an underlying
relational database schema, but a system we wrote ourselves in which data
are stored in objects that belong in classes. This is nevertheless a general
database management system using caches, session control, and a powerful
query language. Typical objects are clones, genes, alleles, papers,
sequences, etc. Each object is stored as a tree, following a hierarchical
structure for the class (called the "model"). Maps are derived from data
stored in tree objects, but precomputed and stored as tables for efficiency.
The system of models allows flexibility and efficiency of storage --missing
data are not stored. A major advantage is that the models can be extended
and refined without invalidating an existing database. Comments can be added
to any node of an object.
Return to List of Questions
------------------------------------------------------------------------
Q1: What is the current version of the ACEDB software?
A1:
The current Unix version is ace.4_5d/e, released 28 August 1997.
Enhancements added during the ACE97 Workshop will be incorporated in the
next release.
The very latest development version is always available at
ftp://ftp.sanger.ac.uk/pub/acedb_src/. This is unsupported and not to be
redistributed.
New! The current Windows 95/NT version is 4.5.4b, 15 May 1998. For the
latest documentation see the WinAce Web page at
http://mendel.medgen.ubc.ca/rbrusk/winace.html.
A Macintosh version is available as version 4.1b1, August 1995.
New! The new (Sanger) webace is at version 2.0a5, March 1998. (See "Can
ACEDB be networked?".)
New! Jade is at version 1.0j, Apr 1998. (See "Can ACEDB be networked?".)
Return to List of Questions
------------------------------------------------------------------------
Q2: Where can I get ACEDB?
A2:
Source code and Unix binaries are available in the following anonymous ftp
sites:
* ncbi.nlm.nih.gov in repository/acedb
* ftp.sanger.ac.uk in pub/acedb
* lirmm.lirmm.fr in pub/acedb
* bioinformatics.weizmann.ac.il in pub/databases/acedb
Linux binaries in ELF and a.out format, from Jeff Bryer:
* ncbi.nlm.nih.gov in repository/acedb
* bioinformatics.weizmann.ac.il in pub/databases/acedb
Windows 95/NT, from Richard Bruskiewich:
* ftp.sanger.ac.uk in pub/acedb/winace
* bioinformatics.weizmann.ac.il in pub/databases/acedb
* Documentation: http://mendel.medgen.ubc.ca/rbrusk/winace.html
MacAce, from Frank Eeckman, Cyrus Harmon and Richard Durbin:
(Note: The authors are not currently able to support MacAce. Latest version
was 4.1b1.)
* ncbi.nlm.nih.gov in repository/acedb/macace
* genome.lbl.gov in pub/macace
* bioinformatics.weizmann.ac.il in pub/databases/acedb
NEC EWS4800 binaries:
* http://www.labs.nec.co.jp/freesoft/freesofte.html
Return to List of Questions
------------------------------------------------------------------------
Q3: What hardware/software do I need to run ACEDB?
A3:
The software is available in binary (pre-compiled) format for a variety of
machines.
* Unix:
o Sun/SunOS 4.x
o Sun/Solaris
o DEC DECstation3100, 5100 etc.
o DEC Alpha/OSF-1
o Silicon Graphics Iris series 4, 5, 6
o IBM RS-6000
o PC 386/486/Pentium with Linux
o NEC EWS4800
o NeXT: contact Patrick Phillips at University of Texas, NeXTmail:
patrick at wbar.uta.edu email: phil at decster.uta.edu
o There exist, or have existed, ports onto Alliant, Hewlett-
Packard, Convex. You may have to contact the developer responsible
for the port to make these real.
* Windows 95/NT
* Macintosh (not currently supported)
The software is also available as source code, so you may be able to get it
working on any machine.
Memory requirements (from Richard Durbin, aug 97)
The amount of memory you require for ACEDB depends very much on how big the
database is (i.e. the disk space used by the database/ subdirectory). Our
rule of thumb is that one typically uses 5-10Mb plus up to 10% of the disk
space size of the database. So with a 200Mb database perhaps 25Mb memory,
and with a 500Mb database (e.g. the C. elegans one) up to 50-60Mb. In fact
for short sessions less memory is used -- it is only when all classes are
explored, or for example when parsing big files that these amounts of memory
get used.
Return to List of Questions
------------------------------------------------------------------------
Q4: Can ACEDB be networked?
A4:
ACEDB Client / Server Computing (from Doug Bigwood, aug97)
There are several client/server models for ACEDB computing and several more
are in development. The start of the ACEDB client/server age began with the
inclusion of aceclient and aceserver in version 4.0. These are C - based and
use the RPC protocol for communication. These executables can be made from
the standard ACEDB distributions.
Starting in version 4.5 an xaceclient is also included with ACEDB.
Xaceclient provides remote read/write access to an aceserver while providing
the user with the same X displays that are found in xace. To use it, you
create an empty database with the appropriate models and start xaceclient.
It will automatically retrieve data from the server declared in
wspec/server.wrm (the Montpellier server in the distribution server.wrm).
The data will be saved locally and can then be viewed with a normal xace.
A perl extension which provides aceclient functionality to Perl 5.x was
developed at ACE95. The files necessary for this perl extension are now
(ACEDB 4.5 and later) included in the wrpc directory of the ACEDB directory
hierarchy. Documentation about how to extend perl is found at
http://probe.nalusda.gov:8000/ace97/perlace/perlacecl.html.
WWWAce and its successor webace were developed to provide a World Wide Web
interface for ACEDB. Webace instructions can be found at
http://probe.nal.usda.gov:8000/acedocs/webace.html, and
http://probe.nal.usda.gov:8000/ace97/webace.html and the program itself at
ftp://probe.nalusda.gov/pub/tools/webace.tar.gz.
A Java-based client called Jade allows communication via sockets to an
aceserver. Jade installation instructions and information on downloading can
be found at http://probe.nalusda.gov:8000/ace97/Jade.installation.html.
There are now development efforts underway to provide additional
client/server functionality to ACEDB including a CORBA server and
socket-based communications. These will likely be included in future
versions of ACEDB. A new C library interface to ACEDB internals will greatly
ease the development of new clients and servers that will support additional
protocols.
Subsequent developments (from Dave Matthews, jun98)
New! A new version of webace, sometimes called webace2, has been developed
at the Sanger Centre. It makes use of the new gifaceserver instead of
aceserver to improve interactive response of the graphical displays,
Javascript, Java, and a new Aceclient.pm module which can be installed into
Perl without recompiling. It also supports the ACEDB ?URL class. The home
page, with links to databases that are using it, documentation, and
information about joining the webace mailing list, is at
http://webace.sanger.ac.uk/.
New! AcePerl, from Lincoln Stein, is an object-oriented Perl interface to
ACEDB. It can connect to remote ACEDB databases, perform queries, fetch ACE
objects, and update databases. The programmer's API is compatible with the
Jade Java API. Home page at http://stein.cshl.org/AcePerl/.
New! Jade and its documentation and demonstration databases have been
updated. Its home page is now http://stein.cshl.org/jade/.
Return to List of Questions
------------------------------------------------------------------------
Q5: What documentation exists for ACEDB?
A5: