Related Experiment Videos
Predicting logP of pesticides using different software.
E Benfenati1, G Gini, N Piclin
1Department of Environmental Health Sciences, Istituto di Ricerche Farmacologiche Mario Negri, Laboratory of Environmental Chemistry and Toxicology, via Eritrea 62, 20157 Milan, Italy. benfenati@marionegri.it
Chemosphere
|September 27, 2003
Summary
This study evaluated logP prediction software for pesticides, comparing calculated values against experimental data from multiple sources. Pallas and KowWin demonstrated the best predictive performance, offering valuable insights for pesticide research.
Area of Science:
- Computational chemistry
- Environmental science
- Agrochemical research
Background:
- LogP values are crucial for understanding pesticide behavior and environmental fate.
- Pesticides present unique challenges for logP prediction due to their structural complexity.
- Existing studies often rely on single experimental datasets, potentially introducing bias.
Purpose of the Study:
- To compare the accuracy of different software in calculating logP values for pesticides.
- To evaluate logP predictors using multiple experimental datasets, accounting for variability.
- To identify the most reliable logP prediction tools for agrochemical applications.
Main Methods:
- Calculated logP values using HyperChem, Pallas, KowWin, and TOPKAT software.
- Compared calculated values against experimental logP data from 235 pesticides across four databases.
- Employed a cross-comparison approach utilizing diverse experimental datasets to mitigate bias.
Main Results:
- Pallas and KowWin exhibited the highest accuracy in predicting pesticide logP values.
- TOPKAT showed moderate predictive performance.
- The cross-comparison method highlighted the importance of using multiple experimental sources.
Conclusions:
- Pallas and KowWin are recommended for accurate logP prediction in pesticide research.
- Utilizing multiple experimental datasets enhances the reliability of logP predictor evaluations.
- The study provides a robust validation of computational methods for complex organic molecules like pesticides.