Is there a need for a Calculated Structures Database ?
- From: Gijs Schaftenaar <schaft-0at0-caos.kun.nl>
- Subject: Is there a need for a Calculated Structures Database
?
- Date: Tue, 11 Aug 1998 13:12:21 +0200 (MDT)
A Calculated Structures Database.
---------------------------------
Not more than perhaps a decade ago, Computational Chemistry used to be the
realm of Quantum Chemists with access to a supercomputer. Computer resources
usually restricted calculations to small molecules. Nowadays this situation
has changed drastically. With the continuous speedup of computer hardware,
calculations on chemically interesting systems have become tractable to the
bench chemist and the study of short lived intermediates and transition
states, often inaccessible to experiment, have become common practice.
The Computational Chemistry community has steadily grown and broadened
and does no longer exist of pure specialists only. The Computational
Chemistry List (CCL; URL: http://www.ccl.net/ccl/welcome.html), a Web
site and an e-mail exploder which allows computational chemistry researchers
from around the world to exchange information and experiences is an important
commucation channel.
With so many structures being calculated, the need for a (comprehensive)
database of calculated structures has been voiced a number of times on the
Computational Chemistry List. The Carnegie-Mellon Quantum Chemistry Archive,
created by Prof. Pople's group for users of the Gaussian package, has been
an attempt to set up such a database. However, for a number of reasons this
database has never been very successful. Gaussian users could (can) contribute
to this database by sending a routinely created archive file by surface
mail to the curators of the database, the Chemistry Publishing Unit of
Carnegie-Mellon University. Access to the database can only be obtained by
purchasing it together with its dedicated software. But for a database to
be successful, ease of contributing to it and ease of data retrieval are
determing factors.
The CAOS/CAMM Center, the Dutch national academic Center for the
Application of Information Technology in Chemistry, is exploring the
possibility to create, curate and maintain a comprehensive archive of
calculated structures at both the ab-initio and semi-empirical level with
emphasis on co-ordinate data. Contributions to this database should be
submitable via e-mail and/or the World Wide Web. We intend to automate
submission by incorporating a submission form into our popular quantum
chemistry visualisation tool MOLDEN, which interfaces not only to Gaussian
but to a range of other Ab Initio and semi-empirical codes. MOLDEN currently
has a distribution of over 800 copies world wide and is freely available to
academic researchers making (coordinate) data contibution an easy task.
Access to the database will be given via the World Wide Web and added value
is intended by linking it to other chemically oriented databases, in
particular to the Quantum Chemistry Literature Database (QCLDB).
The Gaussian archive file format might serve as a starting point for the
definition of a new and richer format which would include semi-empirical
calculations and links to literature and other chemical databases. Because
the Center maintains the MOLDEN software, adaptions to deal with a new
data format can easily be implemented and distributed.
With this message we try to measure the potential (mental) support for
such an initiative and to determine the willingness among computational
chemists to contribute data to the intended database of calculated structures.
This will help us to determine if the allocation of computational and
human resources to this effort can meet with sufficient enthousiasm to
start. For the database to reach the critical mass needed to become a
successful operation, in addition to the processing of existing data sources,
the active contribution of data from many private sources is necessary.
Therefor your answer to the following questions is very much appreciated.
1. Do you think that the proposed initiative is advisable and/or needed ?
2. Would you be willing to contribute data to the proposed database, using
either a MOLDEN form or a WEB based form?
Your answers are welcomed at the e-mail addresses:
schaft-0at0-caos.kun.nl (Gijs Schaftenaar, Computational Chemist and MOLDEN
author CAOS/CAMM Center.)
or
noordik-0at0-caos.kun.nl (Jan Noordik, Director CAOS/CAMM Center)
------------------+-----------------------+------------------------------
CAOS/CAMM Center Phone: +31(0)24-3653386
University of Nijmegen +31(0)24-3653391(secr.)
Toernooiveld 1 Fax: +31(0)24-3652977
6525 ED Nijmegen e-mail: post-0at0-caos.kun.nl
The Netherlands URL: http://www.caos.kun.nl
------------------+-----------------------+------------------------------
The CAOS/CAMM Center is the Dutch Academic Facility for the Application of
Information Technology in:
Organic Chemistry: CAOS
Computational Chemistry, Molecular Modeling, 3D Databases: CAMM
Bioinformatics: CAMMSA
It is also the Dutch National Node in EMBnet and GDB.
------------------+-----------------------+------------------------------