Advances in computer-aided drug design: From molecular docking to artificial intelligence-driven therapeutic discovery

Computer-aided drug design (CADD) has changed the way the pharmaceutical industry works because of the use of computational tools to allow drug discovery and development to be more efficient and effective. Here, we provide an overview of the main methods, technologies, and strategies used for CADD, such as structure-based drug design, ligand-based drug design, molecular docking, virtual screening, molecular dynamics simulations, and quantitative structure–activity relationships. Some of the benefits of advanced docking algorithms and virtual screening software tools (eg, AutoDock, Glide, and PyRx) for evaluating high-affinity ligands and lead compound optimization will also be discussed. Furthermore, the paper describes inverse docking for target identification and drug repurposing, as well as the combination of absorption, distribution, metabolism, excretion, and toxicity prediction tools to investigate pharmacokinetics and toxicity in silico. The review also highlights molecular dynamics for measuring protein-ligand stability and artificial intelligence, whereby machine learning and deep learning have greatly varied the speed of drug repurposing, hit-to-lead optimization, and target identification. Through case studies and software comparisons, this review demonstrates how the combination of in silico computational models and experimental data is reducing drug discovery costs and timeframes, while delivering promising novel therapeutics. Lastly, the incorporation of artificial intelligence is clearly a step forward, elevating CADD as a modern requirement of drug discovery. Significance Statement: Computer-aided drug design is rapidly evolving from classical docking and quantitative structure–activity relationships toward integrated, artificial intelligence-enabled workflows that can reduce discovery time and cost while improving decision quality. This review synthesizes practical strategies, common software platforms, and best-practice integration of docking, virtual screening, molecular dynamics, absorption, distribution, metabolism, excretion, and toxicity, and machine learning to help a broad readership understand how modern in silico pipelines accelerate hit identification, optimization, and repurposing.