Scaffold Optimization Strategies in Medicinal Chemistry: Chemical Approaches and Artificial Intelligence Applications

Chemical skeleton optimization is a core strategy in medicinal chemistry. By adjusting the chemical skeleton to improve biological activity, selectivity and ADMET properties, it can enhance the drugability of drug candidate molecules. In chemical methods, such as skeleton transitions, heterocyclic substitution, and skeleton editing have been applied to enhance efficacy, selectivity, and pharmacokinetic properties. These strategies allow researchers to explore new chemical space, overcome limitations of existing leads, and develop molecules with better safety profiles. In recent years, AI technology has completely transformed scaffold design by deeply integrating skeleton transition algorithms with reinforcement learning frameworks to achieve data-driven molecular representation, generative modeling, and structure-activity prediction. This article will describe the optimization strategies of chemical skeletons, the application of AI in skeleton transitions, case analyses, future trends and challenges. The synergy between chemical expertise and artificial intelligence is expected to accelerate drug discovery and increase the success rate of new therapy development.