Unraveling mercury bioavailability: From molecular mechanisms to AI-driven environmental solutions

Mercury (Hg) bioavailability is a central determinant of its environmental behavior, transformation into toxic methylmercury (MeHg), and subsequent accumulation across biological systems. Despite extensive research, inconsistencies in definitions and methodological approaches have hindered the integration of Hg bioavailability assessments across microorganisms, plants, animals, and humans. This review provides a comprehensive synthesis of the current knowledge on Hg bioavailability by: (1) summarizing its conceptual framework and mechanistic pathways across multiple biological levels; (2) critically evaluating key analytical techniques-Diffusive Gradient in Thin Films (DGT), whole-cell biosensors, and isotope tracing-highlighting their respective advantages and limitations; and (3) focusing on the bioavailability and exposure risks of MeHg in rice, a critical pathway for human intake in many regions. Furthermore, we present recent advances in artificial intelligence (AI), particularly machine learning (ML), as powerful tools to assess Hg bioavailability across biological hierarchies. ML models integrating geochemical, biological, and environmental data offer interpretable, high-resolution predictions of Hg bioavailability in microbial systems, ecological food webs, and human exposure scenarios. By integrating mechanistic insights with emerging AI-driven modeling strategies, this review bridges the gap between experimental understanding and scalable prediction of Hg bioavailability, providing a foundation for precision monitoring, risk assessment, and policy development in support of global mercury control efforts.