Computational Prioritization of Multi-Target Antidepressant Analogs via Hierarchical Virtual Screening

Major depressive disorder (MDD) treatment is limited by high rates of incomplete symptom remission, necessitating polypharmacological approaches. This study executes a hierarchical in silico workflow as a strictly predictive, hypothesis-generating stage to identify putative multi-target antidepressant candidates. A ligand-based similarity screen of the ZINC database generated an initial library of 14,561 analogs based on the structural templates of established antidepressant drugs. This dataset was subjected to multi-stage computational filtering integrating quantitative structure-activity relationship (QSAR) modeling, molecular docking, and predictive ADMET profiling. The pipeline prioritized 298 compounds that demonstrate in silico binding scores potentially higher to their respective parent reference drugs across a panel of monoaminergic transporters, serotonergic receptors, and glutamatergic targets. These prioritized analogs exhibit computationally optimized multi-target profiles, maintaining predicted blood-brain barrier permeability while reducing in silico liabilities for hERG channel inhibition and hepatotoxicity.