Validation

OpenTox is developing unified and objective validation routines for model and algorithm developers and for external (Q)SAR programs, including procedures for validation with artificial test sets (e.g. n-fold cross-validation, leave-one-out, simple training/test set splits) as well as facilities to validate against confidential data. 

An important goal is to integrate statistical tests for the comparison of (Q)SAR models under consideration of their applicability domain, a versioned database to store validation results and their history, and tools for the inspection of the toxicological plausibility of (Q)SAR predictions.

These validation techniques will be available for all (Q)SAR models (OpenTox and external programs) that are plugged into the framework. This will help to compare algorithms and (Q)SAR models objectively and to speed up the development cycle.

The OECD Guidelines for (Q)SAR Validation will be followed.

Interfaces: OpenTox Interface for Validation REST Web Service is available in the Interfaces section.

Components: Current OpenTox Components supporting Validation is available in the Components section.