Artificial Neural Network (ANN) Optimization With MoGEO, Molecular Docking, Pharmacokinetic Studies and DFT Studies of Ruthenium (III) and Vanadium (V) Complexes

ABSTRACT In this study, four metal complexes [VL (phen)], [RuL (phen)], [VL (bpy)] and [RuL (bpy)]—were synthesized and structurally characterized using comprehensive analytical and spectral techniques. Density functional theory (DFT) calculations (B3LYP) were employed to examine their electronic properties and reactive sites, while molecular docking studies revealed strong hydrogen‐bonding interactions of the furan, thiazole, bipyridine, and phenanthroline moieties with Mpox and COVID virus receptor proteins. Key metal–ligand interactions, notably oxygen coordinated to vanadium and chlorine coordinated to ruthenium, exhibited high binding affinity. In silico ADMET profiling indicated drug‐like properties but also highlighted potential limitations in oral bioavailability due to high molecular weight, lipophilicity and low solubility. To enhance predictive modelling, a molecular gradient evolution optimizer (MoGEO)—driven artificial neural network (ANN) framework was developed, integrating molecular energy shifts and docking interaction data into the learning process. The ANN‐MoGEO approach achieved superior accuracy in predicting synthesis yields, docking scores and spectral peaks, outperforming conventional optimizers. These findings not only deepen the understanding of Ru (III) and V(V) complexes but also demonstrates ANN‐MoGEO's potential as a powerful tool for molecular property prediction in drug design.