Home- Research
- Conferences
- Program & Schedule
- Location
- KM in R&D Workshop
- Sponsors & Exhibitors
- Call for Contributions
- Bursary Awards
- Poster Session
- Program 2007
- Knowledge Management Workshop
- ADMET
- Beger, R
- Cherkasov, A
- Contrera, J
- Helma, C
- Hopfinger, A
- Jamois, E
- Klon, A
- Madden, J
- Pelletier, D
- Poroikov, V
- Richard, A
- Tomaszewski, J
- Program 2006
- Photo Gallery
- Workshops & Training
- Program
- Exhibition
- Registration
- About
- Overview
- Contact
- Support
- Schedule
|
|
|
|
|
|
|
|
|
| About Christoph Helma (University of Freiburg and in silico toxicology) |
|
Dr. Christoph Helma is a computational toxicologist with more than 10 years experience in the development and application of advanced data mining techniques for toxicological applications, predominantly for the identification of Structure-Activity Relationships. He is the developer of the award winning lazar system for the prediction of toxic activities (www.predictive-toxicology.org/lazar/), has been an invited keynote speaker at major scientific conferences and has published more than 30 peer-reviewed articles. C.H. was the main organizer of the Predictive Toxicology Challenge 2000-2001, editor for a special section in Bioinformatics, and editor of a book about Predictive Toxicology. He is the founder and head of "in silico toxicology", a spin-off company of the University Freiburg, Germany. He is presently the bioinformatics workpackage leader for the Sens-it-iv EU project and works on the implementation of a inductive database for the interactive analysis of toxicogenomics, -proteomics and -metabonomics data. He was awarded with the Research Prize for Alternative Methods to Animal Experiments (German Federal Ministry on Consumer Protection, Food and Agriculture, 2005) and the Research Prize for Cancer Research without Animal Experiments (Doctors Against Animal Experiments, 2006).
|
|
In-silico prediction of chemical toxicity: Lazy-Structure-Activity-Relationships (lazar) and the OpenTox framework
Christoph Helma, University of Freiburg and in silico toxicology, Germany
This presentation gives an overview of the lazar (Lazy Structure Activity Relationships) system for the prediction of biological activities and its application for the prediction of carcinogenicity, an endpoint that is very hard to predict with existing (Q)SAR techniques.
lazar uses a modified k-nearest-neighbor algorithm, that is capable of detecting activity-specific chemical similarities, to derive predictions for untested structures from a database with experimental toxicities. lazar relies on relatively few model assumptions and provides the rationales for predictions in an understandable and traceable manner. The system is capable of discriminating reliably between trustworthy and untrustworthy predictions (e.g. for structures that fall beyond the scope of the training set) by assigning a confidence index to each prediction.
Cross-validation experiments with various carcinogenicity and mutagenicity endpoints from the Carcinogenic Potency Database (1376 compounds) gave predictive accuracies between 80 and 90 % for structures within the applicability domain of the training data. The confidence index correlates well with predictive accuracies, which indicates that lazar can reliably identify cases where the information in the database is insufficient and/or contradictory to derive valid predictions.
The presentation will finish with a brief presentation of the OpenTox proposal. It suggests the development of a framework that provides unified access to toxicological (Q)SAR models, algorithms and data as well as to information supporting the toxicological interpretation of (Q)SAR predictions.
A public interface for the lazar system and the source code for the program is freely available at the website http://www.predictive-toxicology.org/lazar/. OpenTox will be an open source project with a public demonstration website.
|
|
|
|
|
|
|
|
|
|
|
