Integrating metabolomics and metagenomics to unravel asthma pathophysiology using artificial intelligence and systems biology

Asthma is a heterogeneous chronic airway disease driven by complex interactions among genetic, environmental, metabolic, and microbial factors. Conventional diagnostic and therapeutic approaches often fail to capture its molecular diversity, underscoring the need for integrative strategies. Metabolomics and metagenomics have emerged as transformative tools for dissecting asthma pathophysiology, offering complementary insights into host biochemical alterations and microbiome-mediated immune modulation. Metabolomic studies across biofluids such as urine, plasma, exhaled breath condensate, sputum, and bronchoalveolar lavage fluid have identified alterations in lipid, amino acid, and short-chain fatty acid metabolism associated with asthma severity and inflammatory endotypes. Metagenomic analyses have further revealed microbial dysbiosis in asthma, including increased airway abundance of Haemophilus , Moraxella , and Neisseria , along with reduced commensal taxa. Early-life gut microbiome changes, such as reduced diversity and increased fungal taxa like Candida and Pichia , have also been linked to increased asthma risk. Recent advances in systems biology and artificial intelligence (AI) have enabled integration of multi-omics datasets to uncover host–microbe–metabolite networks that drive asthma heterogeneity. Machine learning and network-based approaches have enabled the accurate identification of predictive biomarkers, delineation of disease endotypes, and prioritization of potential therapeutic targets. Despite these advances, the widespread implementation of AI-driven multi-omics approaches in asthma is challenged by algorithmic biases, data security and privacy concerns, and the black-box nature and limited interpretability of these models, necessitating rigorous clinical validation before their translation into clinical practice. This review provides a comprehensive overview of metabolomic and metagenomic alterations in asthma, explores their convergence through microbial–metabolic crosstalk, and highlights the potential of AI-guided systems biology approaches to advance next-generation precision respiratory medicine.