Harnessing Topological Indices in QSPR: Predicting Molecular Properties for Parkinson's Disease Therapeutics

Abstract

Topological indices are numerical invariants derived from molecular graphs play a crucial role in Quantitative Structure-Property Relationship (QSPR) modeling, helping in the prediction of molecular properties relevant to drug discovery. In this study, we predicted eight physicochemical characteristics (PC-C) of thirteen Parkinson’s illness medicines through QSPR analysis and topological indices. We computed these indices using Python-based tools (RDKit, NetworkX) and performed regression analysis using Scikit-learn and SciPy. The results show that how some topological indices exhibit strong correlations with several PC-C, demonstrating their value in drug development pipelines.