Supramolecular inclusion complexes are widely used in drug delivery and other fields, with the advantages of controllable structures, high stability, excellent biocompatibility, and the ability to improve drug solubility and achieve controlled release. However, traditional screening methods rely on experimental trial and error, which suffer from long cycles, high costs, and low throughput, limiting research and development efficiency. In recent years, the development of artificial intelligence has provided new solutions for the screening of inclusion complexes. This paper systematically reviewed the core technological system of AI in the screening of inclusion complexes, focusing on two aspects: prediction and optimization of key properties and rational design of host molecules, summarizing their specific application progress. At the same time, we analyzed the current core challenges, including data scarcity, insufficient model interpretability, and limited generalization capabilities, and propose future development directions such as building standardized databases, integrating physicochemical principles (e.g., molecular mechanics and thermodynamics), and establishing closed-loop research and development platforms. This review aims to provide a systematic reference for the in-depth application of artificial intelligence in the field of supramolecular inclusion complexes.
Deng et al. (Sat,) studied this question.