Integrating Metabolomics and Machine Learning for Advanced Chemical Detection

Metabolomics has emerged as a powerful analytical approach for comprehensive chemical profiling in complex biological and environmental systems. The increasing volume, dimensionality, and complexity of metabolomics data have driven the adoption of machine learning (ML) techniques to enhance chemical detection, classification, and interpretation. This narrative review critically discusses the integration of metabolomics and machine learning for advanced chemical detection, with particular emphasis on analytical workflows, data preprocessing strategies, supervised and unsupervised learning models, and validation approaches. In this context, advanced chemical detection refers to the data-driven identification, classification, and quantification of chemical signatures in complex matrices with improved sensitivity, selectivity, robustness, and interpretability. Current applications across food science, environmental monitoring, clinical diagnostics, and exposomics are discussed, along with key challenges related to data quality, interpretability, and reproducibility. Finally, future perspectives on explainable AI, multimodal data integration, and standardized pipelines are highlighted.