Chemistry Reference
In-Depth Information
(http://apps.ideaconsult.net:8080/ToxPredict) and CAESAR (http://www.caesar-
project.eu/software/index.php). LAZAR (Lazy Structure-Activity Relationships)
is an experimental version based on OpenTox, whose linear fragments of the
query structure or in one of the training structures are determined exhaustively by
molecular feature algorithm (MOLFEA). CAESAR is based on WEKA
algorithms [53-55].
WEKA (Waikato Environment for Knowledge Analysis) is a free software
available under the GPL (General Public License), developed at the University of
Waikato, New Zealand (Fig.
2
). It has a collection of machine learning algorithms
for data mining tasks [56]. MOLFEA (Molecular Feature Miner) is a domain
specific inductive database which mines for fragments in chemicals. The
compounds are stored in the SMILES (Simplified Molecular Input Line Entry
System) format and fragments are formulated in the SMARTS language [57].
Figure 2:
The Weka Software. Weka is a collection of machine learning algorithms for data
mining tasks (http://www.cs.waikato.ac.nz/ml/weka/).
