Abstract

BackgroundLower respiratory tract infections (LRTIs) exacerbate chronic airway diseases, yet phenotype-specific microbial signatures are poorly defined. We applied broncho-alveolar lavage fluid (BALF) genomic sequencing to identify biomarkers for asthma and chronic obstructive pulmonary disease (COPD).MethodsBetween December 2023 and February 2025, 1–146 adults with suspected LRTI enrolled from the First Hospital of Jilin University underwent BALF next-generation sequencing. Patients were stratified by lung function, with the impaired pulmonary function group further divided into asthma, COPD-mild-moderate, and COPD-severe subgroups. Disease-specific key biomarkers were identified using machine learning algorithms and analyzed for co-occurrence.ResultsImpaired pulmonary function was not only associated with pathogenic microorganisms and its higher microbial burden, but also associated with a distinct community structure. Random forest models revealed disease-specific biomarkers, with Prevotella intermedia, Finegoldia magna, and Human parvovirus enriched in asthma, Veillonella parvula, Human respiratory syncytial virus, and Haemophilus influenzae enriched in COPD-mild-moderate, and Human respiratory syncytial virus, Human coronavirus, and Human parainfluenza virus enriched in COPD-severe. Co-occurrence network identified hubs linking asthma-centric (Haemophilus parainfluenzae and Schaalia odontolytica) and COPD-centric (Klebsiella pneumoniae, Veillonella parvula, and Streptococcus constellatus) clusters, suggesting potential cross-phenotype microbial crosstalk.ConclusionsGenomic sequencing profiling delineates distinct yet overlapping airway microbiota across separate pulmonary dysfunctional diseases - asthma and COPD. Compact biomarker panels classify each condition accurately and reveal shared microbial hubs that may drive chronic inflammation and exacerbations, supporting microbiome-guided precision diagnostics and therapy.