Abstract
Chronic obstructive pulmonary disease (COPD) is a heterogeneous condition arising from interactions among genetic susceptibility, environmental exposures, altered cellular responses and progressive lung injury. This complexity limits the ability of conventional clinical measures or any single molecular layer to explain differences in disease development, progression and treatment response. This review examines how artificial intelligence (AI)-mediated integration of selected omics including genomics, epigenomics, transcriptomics, single-cell and spatial omics, proteomics, metabolomics, lipidomics and microbiomics could provide a more connected understanding of COPD biology. These complementary layers capture inherited susceptibility, exposure-related regulation, cell-specific dysfunction, active molecular pathways, metabolic alterations and host-microbial interactions. Although previous publications have examined COPD multi-omics, computational integration or AI applications, this review proposes a conceptual, hypothesis-generating framework linking selected omics across disease development and biological mechanisms to potential biomarkers and therapeutic opportunities. Such integration may generate testable hypotheses, identify patient subgroups with biologically meaningful characteristics and help in biomarker development, therapeutic-target identification, drug repurposing, treatment-response prediction and clinical-trial design. However, direct evidence demonstrating the clinical value of AI-integrated COPD omics is warranted. Translation could be further constrained by small and selected cohorts, technical heterogeneity, confounding factors like smoking, treatments, limited population diversity and insufficient independent and prospective validation. By focusing on selected omics layers that are among the most relevant to COPD biology and clinical heterogeneity, this framework identifies timely opportunities for integrative research that could lay the groundwork for developing and testing biomarkers and therapeutic strategies. Progress will require longitudinal multicenter studies, standardized measurements, interpretable and biologically informed AI methods, experimental confirmation and prospective evidence that resulting approaches improve clinical decisions or patient outcomes.