IUBio GIL .. BIOSCI/Bionet News .. Biosequences .. Software .. FTP

ACE4, bugs and description

Jean Thierry-Mieg mieg at watson.wustl.edu
Sun May 28 18:15:46 EST 1995


#############BUGS###############
Leon Avery already found 2 bugs in tace

1) you must ace comment out several lines in options.wrm
as follows

/* use these for synmap 
//  _VMap              -V -D SYN -M STSMAP
//  _VAssembly         -V -D SYN -M DOGMAP
//  _VClone            -V -D SYN -M CloneMAP
*/

2) list option in tace is bugged
Easy to fix but very annoying, I suppose the code was never tested 
after modification at ace95

3) solaris linking is a problem
best is to recompile on your own solaris machine

*******************************************************
############ DESCRIPTION ############################

Ace 4.0 is a very hasty distribution motivated by the 
worm meeting taking place this week, the code nevertheless
contains many interesting advances, which can be described as follows:

A number of people have contributed different more or less independant pieces

Kernel Level:
Overall, the system is faster and uses less memory:
The lists in the main window are 'instantaneous'
The cache manager balanced was modified,
the dictionary system now has 2 levels
  These modifications were motivated by teh IGD project and developped
in Montpellier (D&J Th-M) with fixes from Cambridge (rd & srk)

Acedb now recognise the platform it runs on, 
it automatically byte swaps if needed, 
but you should initialise it on the platform (DEC-SUN-IBM ...) which
you intend to use most. This system was written by Simon Kelley

Table Maker:
The table maker (JTM) now works correctly inside complex objects
this was fixed in 3.5, but not in 2.0

Query Builder
A new version of the query buider interface has been contributed
by Suzana Lewis, replacing an older version also from LBL.

Server/Client:
Indisagreement with Richard, who wanted to transmit compiled byte
streams, we constructed in Montpellier over the last year, with the
help of the Heidelberg group, a client server exchanging ace streams
thru rpc.  This amounts to writing a multiphreaded tace who can
answer and give write access to several clients in parallel. During
the ace95 workshop the system was enhanced to allow configurable
ports and perl exports.  The purpose of this perl interface is the
enhancement of the ace server run by Bigwood at the national
agriculture library. This Web server augments the acedb data with
url to other database before returning the answers.  A server client
architecture is complex to handle and not useful in small labs.

Configurability of the models:
The models have been completely rewritten with the tag2 system introduced
in ace.3.0 This allows the curators to name their classes at will, without
losing the displays, i.e. one may call a gene a locus without penalty.
This modification affects the genetic map and the grid display, so you may have
to modify your personal data. It does not affect the vMap (introduced in 3.0
which was alreay tag2) nor the horizontal physical map which was kept as is.

Configurability of the genetic map display:
Simon Kelley has thorougly rewritten the control of the genetic map
display. New features include: an easier control of what is displayed, 
the possibility to configure the color code, and a simple way to
save and reuse your choices. Comments welcome.

Sequence Map display:
Richard has introduced several new functionalities in the display
of homologies and their scores. Also a possibility to read and write
peptide sequences. 

Blixem:
This sequence algnment tool by Erik Sonnhammer is enhanced since
2.0, it gives an example of a contributed tool, with its own 
user interface, somehow different from the rest of acedb, despite
the fact that it uses the same toolbox

Molecular Pathways
This code, originally written by Stan Letowsky, was adapted and
linked into ace4.0 by Richard. You may need some new data to test
it.

Mosaic Help
A number of people, including Sam Cartinhour, John Barnett and Jaime
Prilusky have modified the acedb help system so that, if possible,
the help will be displayed in a mosaic window, allowing direct
access to the documentation servers.

############## WORM DATA ###############################

The worm dataset distributed on the ncbi server (update.4*)
was collated in Cambridge by Richard and co.


*************************************
############## ACEMBLY #################################
The sequence assembly and edition package Acembly, written to
support shotgun sequencing and study of dna variations, by Ulrich
Sauvage, Danielle and Jean T-M will be distributed independantly in
the near future.  This programs relies on the acedb database and
toolbox but its complex graphic interface is nearly enterely new.

We consider it to be an example of the modularity of acedb. In
acembly, all the new work went into the conception of the display,
with the acedb system considered as a reliable black-box.

***************************************
############## Using ace 4.0 ##############################

Again this is a test release,
Send comments, try it on public datafiles, port your data to
the new format when needed, but I do not yet garantee that the
database kernel is as reliable as 3.7 and will not accidentally
loose the data. 

############################################################

Jean T-M





More information about the Acedb mailing list

Send comments to us at archive@iubioarchive.bio.net