Abstract
Designing effective drug therapies requires balancing competing objectives, such as therapeutic efficacy, safety, and cost efficiency-a task that poses significant challenges for conventional optimization methods. To address this, we propose the multi-objective spider-wasp optimizer (MOSWO), a novel approach uniquely emulating the cooperative predation dynamics between spiders and wasps observed in nature. MOSWO integrates adaptive mechanisms for exploration and exploitation to resolve complex trade-offs in multiobjective drug design. Unlike existing approaches, the algorithm employs a dynamic population-partitioning strategy inspired by predator-prey interactions, enabling efficient Pareto frontier discovery. We validate MOSWO's performance through extensive experiments on synthetic benchmarks and real-world case studies spanning antiviral and antibiotic therapies. Results demonstrate that MOSWO surpasses state-of-the-art methods (NSGA-II, MOEA/D, MOGWO, and MOPSO), achieving 11% higher hypervolume scores, 8% lower inverted generational distance scores, 9% higher spread scores, a 30% faster convergence, and superior robustness against noisy biological datasets. The framework's adaptability to diverse therapeutic scenarios underscores its potential as a transformative tool for computational pharmacology.