Quantum-mechanical QSAR/QSPR descriptors from momentum-space wave functions

Area of Science:

• Quantum Chemistry
• Computational Chemistry
• Molecular Modeling

Background:

• Quantitative Structure-Activity Relationship (QSAR) and Quantitative Structure-Property Relationship (QSPR) methods are widely used for predicting molecular properties.
• These traditional methods often rely on descriptors derived from molecular structure in real space.

Purpose of the Study:

• To introduce and evaluate novel quantum-mechanical descriptors derived from electron momentum distribution functions.
• To assess the predictive power of these descriptors for various molecular activities and properties.
• To compare the performance of these new descriptors against traditional QSAR/QSPR approaches.

Main Methods:

• Calculation of one-dimensional radial distribution functions of electron momentum from ab initio wave functions.
• Derivation of quantum-mechanical descriptors as parameters from these distribution functions.
• Application of the descriptors to predict molecular properties including McGowan's volume, gas-chromatographic retention time, and partition coefficients.

Main Results:

• Quantum-mechanical descriptors derived from electron momentum distribution show high precision in predicting molecular properties.
• The predictive accuracy favorably compares with established QSAR/QSPR methods.
• Successful application to diverse properties such as McGowan's volume, chromatographic retention time, partition coefficients, hyperpolarizability, and narcotic activity.

Conclusions:

• One-dimensional radial electron momentum distribution functions provide valuable quantum-mechanical descriptors.
• These descriptors offer a powerful and precise alternative for molecular property prediction.
• The method demonstrates broad applicability across various chemical and physical molecular characteristics.