Metal-organic frameworks (MOFs) are materials used in important industrial applications such as chemical separations, energy storage, and water harvesting. MOFs contain intricate networks of nano-sized pores. The geometry of the networks determines how molecules move through the MOF and interact with the networks’ pore walls. Computational analysis could help describe the molecular motion, but current methods are limited because they do not capture the complexity of the pore network and chemical reactions that take place inside it. This CAREER project will construct a computational scheme based on a pore graph to quantify complex pore networks and their chemistry. The pore graph will be combined with molecular modeling and machine learning to better understand molecular motions and dynamics in MOFs. The results will accelerate the design of next-generation MOF-based materials for more efficient chemical separations and energy technologies. The project will train students in artificial intelligence (AI) and machine learning (ML). A new chemical engineering course will increase AI awareness among students. Summer workshops and an online course in AI will be created for professional education. Summer camps and a partnership with a local high school will build AI literacy among pre-college students. This CAREER project will develop an integrated framework combining graph theory, molecular modeling, and ML to understand confined phenomena in MOFs where current methods fall short. The core innovation will be the development of the pore graph, a unified mathematical representation that transforms intricate pore networks and chemistry into quantifiable objects. This integrated framework will be applied to address unresolved scientific challenges in three key areas: (1) adsorption thermodynamics, to reveal why larger pores anomalously condense earlier or simultaneously with smaller ones, impacting experimental characterization for MOFs; (2) molecular transport, to