Key points are not available for this paper at this time.
The growing demand for safe, sustainable, and clean-label food preservation strategies has accelerated interest in plant-derived essential oils (EOs) as natural antimicrobial agents. These chemically complex mixtures exhibit broad-spectrum activity against foodborne pathogens and spoilage microorganisms through multitarget mechanisms, including membrane disruption, metabolic interference, oxidative stress induction, and modulation of quorum sensing and biofilm formation. However, their practical application remains constrained by compositional variability, physicochemical instability, matrix-dependent efficacy, and sensory limitations. Recent advances in multiomics technologies have shifted the understanding of EO-microbe interactions from descriptive observations to mechanistic, systems-level insights. Transcriptomic, proteomic, and metabolomic analyses reveal coordinated cellular responses, including stress adaptation, efflux activation, and metabolic reprogramming, providing a foundation for rational optimization of EO-based preservation strategies. In parallel, artificial intelligence (AI) and machine learning (ML) offer complementary tools for modeling complex interactions and predicting antimicrobial outcomes. However, their application remains limited by data heterogeneity, insufficient validation in real food systems, and challenges in interpretability. This review critically synthesized mechanistic, omics-driven, and AI-enabled approaches in EO-based food preservation, while addressing key translational barriers, including lack of standardization, regulatory complexity, and scalability. Integrating mechanistic biology with data-driven modeling provides a pathway toward more reliable and precision-oriented EO preservation systems aligned with sustainable food safety demands.
Bhuiyan et al. (Wed,) studied this question.