QSAR study on toxicity to aqueous organisms using the PI index

Area of Science:

• Medicinal Chemistry
• Computational Chemistry
• Toxicology

Background:

• Benzene derivatives are widely used, necessitating accurate prediction of their physicochemical and toxicological properties.
• Quantitative Structure-Activity Relationships (QSAR) and Quantitative Structure-Toxicity Relationships (QSTRs) are crucial for predicting chemical behavior.
• The Padmakar-Ivan (PI) index is a novel topological descriptor with potential for QSTR modeling.

Purpose of the Study:

• To develop QSTR models for predicting hydrophobicity (logP) and toxicity (pEC50) of benzene derivatives.
• To evaluate the efficacy of the Padmakar-Ivan (PI) index in these QSTR models.
• To compare the predictive performance of the PI index against other topological indices.

Main Methods:

• Development of multiparametric QSTR models incorporating the PI index and indicator parameters.
• Prediction of hydrophobicity (logP) and toxicity (pEC50) for benzene derivatives.
• Validation of model predictive ability using cross-validation techniques.

Main Results:

• Excellent quantitative structure-toxicity relationships were established for both logP and pEC50 of benzene derivatives.
• The PI index, along with specific indicator parameters, proved highly effective in the developed multiparametric models.
• Cross-validation confirmed the strong predictive power of the QSTR models.

Conclusions:

• The PI index is a valuable descriptor for developing robust QSTRs for benzene derivatives.
• Multiparametric models incorporating the PI index offer excellent predictive accuracy for hydrophobicity and toxicity.
• The PI index demonstrates superiority over several other topological indices in this QSTR context.