Modeling and prediction of octanol/water partition coefficient of pesticides using QSPR methods
Abstract
<jats:sec>
<jats:title content-type="abstract-subheading">Purpose</jats:title>
<jats:p>The purpose of this paper is to predict the octanol/water partition coefficient (<jats:italic>K</jats:italic><jats:sub>ow</jats:sub>) of 43 organophosphorous compounds.</jats:p>
</jats:sec>
<jats:sec>
<jats:title content-type="abstract-subheading">Design/methodology/approach</jats:title>
<jats:p>A quantitative structure-property relationship analysis was performed on a series of 43 pesticides using multiple linear regression and support vector machines methods, which correlate the octanol-water partition coefficient (<jats:italic>K</jats:italic><jats:sub>ow</jats:sub>) values of these chemicals to their structural descriptors. At first, the data set was randomly separated into a training set (34 chemicals) and a test set (nine chemicals) for statistical external validation.</jats:p>
</jats:sec>
<jats:sec>
<jats:title content-type="abstract-subheading">Findings</jats:title>
<jats:p>Models with three descriptors were developed using theoretical descriptors as independent variables derived from Dragon software while applying genetic algorithm-variable subset selection procedure.</jats:p>
</jats:sec>
<jats:sec>
<jats:title content-type="abstract-subheading">Originality/value</jats:title>
<jats:p>The robustness and the predictive performance of the proposed linear model were verified using both internal and external statistical validation. One influential point which reinforces the model and an outlier were highlighted.</jats:p>
</jats:sec>
