Convergence of computer-aided drug discovery and artificial intelligence: Towards next-generation therapeutics

Abstract

The field of drug discovery has undergone transformative changes with the rapid advancement of computing technology. Both academia and the pharmaceutical industry are increasingly adopting computational approaches, with computer-aided drug discovery (CADD) enhancing researchers’ ability to develop cost-effective and resource-efficient solutions. Advances in computational power enable us to explore chemical spaces beyond human capabilities, construct extensive compound libraries, and efficiently predict molecular physicochemical properties and biological activities. Furthermore, artificial intelligence (AI) is now deeply integrated throughout the drug discovery process. As an advanced methodology within CADD, AI-driven drug design (AIDD) accelerates critical stages, including target identification, candidate screening, pharmacological evaluation, and quality control. This approach not only shortens development timelines but also reduces research risks and costs. However, translating computational results for small molecules into successful wet-lab experiments often proves more complex than anticipated, and CADD development still faces persistent constraints. With the evolution of AI tools and the maturation of emerging technologies, CADD holds promise for driving deeper transformations in drug development. • Artificial intelligence (AI) enables rapid de novo molecular generation, ultra-large-scale virtual screening, and predictive modeling of absorption, distribution, metabolism, excretion, and toxicity (ADMET). • Hybrid AI-structure/ligand-based virtual screening and deep learning scoring functions significantly enhance hit rates and scaffold diversity. • The combination of public databases and machine learning models helps overcome structural and data limitations for historically undruggable targets. • Integration of AI-driven in silico design, automated robotics for synthesis/validation, and iterative model refinement compresses timelines exponentially.