Related Experiment Videos
Validation of a QSAR model for acute toxicity
M Pavan1, T I Netzeva, A P Worth
1European Chemicals Bureau, Institute for Health and Consumer Protection, Joint Research Centre, European Commission, Via E. Fermil, 21020 Ispra (VA), Italy. manuela.pavan@jrc.it
SAR and QSAR in Environmental Research
|April 29, 2006
Summary
A new quantitative structure-activity relationship (QSAR) model predicts fathead minnow acute toxicity. Statistical validation and domain definition ensure reliable predictions for untested organic chemicals, aiding regulatory assessment.
Area of Science:
- Environmental toxicology
- Computational chemistry
- Ecotoxicology
Background:
- Quantitative structure-activity relationship (QSAR) models are crucial for predicting chemical toxicity.
- Validating QSAR models requires rigorous statistical assessment and clear definition of their applicability domain.
- Predicting aquatic toxicity is essential for environmental risk assessment and chemical regulation.
Purpose of the Study:
- To develop and validate a QSAR model for predicting the acute toxicity of chemicals to fathead minnow (Pimephales promelas).
- To demonstrate the importance of statistical validation and applicability domain definition for QSAR model reliability.
- To provide a tool for predicting toxicity of untested organic chemicals for regulatory purposes.
Main Methods:
- Development of a QSAR model using multivariate linear regression (MLR) and Genetic Algorithm-Variable Subset Selection (GA-VSS).
- Utilized a dataset of 408 heterogeneous chemicals and theoretical molecular descriptors.
- Performed external validation using OECD Screening Information Data Set (SIDS) data for 177 High Production Volume (HPV) chemicals.
Main Results:
- Achieved good predictivity (=72.1) in external validation using OECD SIDS data.
- The developed QSAR model complied with all five OECD principles for QSAR validation.
- Demonstrated the necessity of statistical validation and domain definition for reliable QSAR predictions.
Conclusions:
- The validated QSAR model is suitable for predicting fathead minnow acute toxicity.
- The model can be used to fill data gaps for untested chemicals within its defined domain.
- This QSAR model supports regulatory chemical assessment and prioritization for aquatic toxicity testing.