* Filename : $RCSfile: intro.phtml,v $
* Purpose : Intro page for the SONNIA Online pages
* Language : HTML
* Authors : L. Terfloth
* Version : $Revision: 1.4 $
* $Date: 2001/04/05 16:23:12 $
* $Author: lothart $
* Copyright: (c) 1999-2001 Prof. Dr. J. Gasteiger, Univ. of Erlangen-Nürnberg
$lastChange = "12.06.2001";
open_document("SONNIA Program Package");
$counterFile = "$weblogDir/KMAPCounterHome.dat";
* $COUNTERFILE = "$weblogDir/KMAPCounterHome.dat";
* schaltet log aus
print_title("Description", $font2, 2, 0);
SONNIA is a self-organizing neural network for the analysis and visualization of data.
It allows both unsupervised (Kohonen) and supervised learning (counterpropagation network)
and provides many techniques for the visualization of chemical data.
SONNIA has the following features:
print_title("Applications", $font2, 2, 0);
SONNIA has shown to be applicable in the following fields
- general data analysis tool
- unsupervised and supervised learning
- projection of data from high-dimensional spaces into 2D planes
- classification and clustering of objects
- modeling and prediction of complex relationships
- planar and toroidal topology of the basis surface
- visualization of chemical structures
/* print_title("Online-Access", $font2, 2, 0);
- analysis of multi-dimensional data
- comparison of compound libraries
- analysis of the similarity and diversity of combinatorial libraries
- classification and prediction of biological activity
- analysis of data from high-throughput screening
- classification of samples from chemical analysis
- simulation of infrared spectra
- reaction classification and clustering
- knowledge extraction from reaction databases
- selection of chemical descriptors
Read a short introduction or go directly to the input forms
(use buttons above):
It is possible to run the calculations for a molecule drawn by an interactive molecule editor
to get all properties calculated by PETRA for datasets of up to $uploadLimit molecules via upload of a datafile
print_title("Availability", $font2, 2, 0);
SONNIA is available for various Unix derivatives:
Please contact Prof. Gasteiger if
you are interested in evaluation of the program.
/* and our web interface is
* not sufficient for your purpose.
/* print_title("Development", $font2, 2, 0);
The current work focusses on the development of a QSPR model for the prediction of pKa values that will be
available within the next version 3.1 to be released end of summer 2001.
- SunOS5.6 and 5.8
The approach we will integrate in PETRA bases on the quantification of the physicochemical effects that have an
influence on the pKa and the derivation of a quantitative model using a feed forward neural network with
a back-propagation learning algorithm.
In an initial model we have shown the high predictive power of this approach as
presented at the \"New Approaches in Drug Design and Discovery\"-Workshop,
Rauischholzhausen March 2001.
print_title("Publications", $font2, 2, 0);
J. Zupan, J. Gasteiger,
\"Neural Networks in Chemistry and Drug Desgin\",
Second Edition, Wiley-VCH, Weinheim, 1999, ISBN 3-527-29779-0.
S. Anzali, J. Gasteiger, U. Holzgrabe, J. Polanski, J. Sadowski,
A. Teckentrup, M. Wagener,
\"The Use of Self-Organizing Neural Networks in Drug Design.\"
In \"3D QSAR in Drug Design -Volume 2\";
H. Kubinyi, G. Folkers, Y.C. Martin (Eds.)
Kluwer/ESCOM, Dordrecht, NL, 1998, pp 273-299.
/* J. Gasteiger, \"Empirical Methods for the Calculation of Physicochemical Data of Organic Compounds\",
in: \"Physical Property Prediction in Organic Chemistry\",
Ed. C. Jochum, M. G. Hicks, J. Sunkel, Springer Verlag, Heidelberg, 119-138, 1988
See a comprehensive list of publications about the calculation of
in organic compounds.
print_title("Manual", $font2, 2, 0);
The manual is online available in HTML,
or downloadable as PDF-File.
print_title("Slide Show", $font2, 2, 0);
print_title("Contact", $font2, 2, 0);
If you are interested in SONNIA please contact:
1. Prof. Johann Gasteiger
2. Dr. Lothar Terfloth
print_footer($lastChange, $KMAPHomeUrl, $counterFile, "home");