QSAR (Quantitative Structure-Activity Relationship) Modeling for Predicting Octanol-Water Partition Coefficients (Log P) Of Benzene Derivatives Using Topological Connectivity Indices and Computational Chemoinformatic Tools.

The Octanol-Water partition coefficient (log P) is a crucial physicochemical parameter that reflects the hydrophobicity of organic compounds. It plays a significant role in fields such as drug design, environmental chemistry, and molecular modeling. However, experimental determination of log P is often time-consuming, labor-intensive, and costly. In this study, a computational approach was employed to predict the log P values of a series of 21 benzene derivatives using molecular connectivity indices. These indices are topological descriptors that encode structural information based on atom connectivity; these were calculated using the Molecular Modeling Pro software. The collected dataset included compounds with known experimental log P values and corresponding molecular connectivity indices. A simple linear regression (SLR) model was constructed with log P as the dependent variable and the connectivity indices as independent variables. The model exhibited a moderate degree of correlation (R² ≈ 0.802) between the predicted and experimental values, indicating robust predictive performance. This research demonstrates the utility of molecular connectivity indices in accurately estimating log P, offering an efficient and cost-effective alternative to traditional experimental methods. This approach can be further extended to other compound classes or enhanced with additional descriptors for improved predictive accuracy.