QSPR analysis of the drugs used to treat major depressive disorder using degree and neighborhood degree based indices.

Introduction: The high risk of suicide among major depressive disorder (MDD) patients, together with its substantial contribution to the global burden of disease, emphasizes the necessity of effective drug evaluation strategies. However, traditional clinical trials for MDD drugs are costly and time-consuming, emphasizing the importance of computational approaches such as QSPR modeling for predicting drug behavior and physicochemical properties. Methods: This study employs degree- and neighborhood degree-based topological indices to characterize the molecular structures of MDD drugs. Topological indices were calculated and analyzed against the physicochemical properties of the drugs using linear, quadratic, and logarithmic regression models to identify the most suitable predictive models, followed by validation to assess their robustness and predictive reliability. Results: Degree- and neighborhood degree-based topological indices exhibited strong correlations with the physicochemical properties of MDD drugs. Among the linear, quadratic, and logarithmic regression analyses performed, the most suitable predictive models were identified, and model validation confirmed their robustness and predictive reliability. Discussion: The strong correlations observed between degree- and neighborhood degree-based topological indices and the physicochemical properties of MDD drugs demonstrate the effectiveness of graph-theoretical descriptors in QSPR modeling. The validated predictive models demonstrate the potential of computational approaches to support efficient drug evaluation and rational drug design, reducing reliance on costly and time-consuming experimental studies.