Athletes exhibit distinct gut microbial signatures and predicted metabolic functions compared with sedentary adults

Background Regular exercise is increasingly recognized as a key modulator of the gut microbiota; however, the comprehensive gut microbial signatures and their potential clinical implications associated with long-term exercise training remain insufficiently characterized, particularly in Asian competitive athletes. This study aimed to compare gut microbial composition and predicted metabolic functions between athletes and sedentary adults and to evaluate the discriminatory value of gut microbial signatures for long-term exercise training. Methods We conducted a cross-sectional study involving 259 participants, including 129 athletes and 130 sedentary individuals. Demographic characteristics, dietary information, and fecal samples were collected. Gut microbial profiles were characterized using 16S rRNA gene sequencing. Microbial diversity, Gut Microbiome Health Index (GMHI), Microbial Dysbiosis Index (MDI), predicted functional pathways, and machine learning models were applied to identify exercise-associated microbial signatures. Results Compared with sedentary individuals, athletes exhibited significantly higher microbial richness and diversity indices ( P  < 0.05), a higher GMHI, and a lower MDI ( P  < 0.001), indicating a healthier gut microbial ecosystem. Taxonomic analysis demonstrated enrichment of beneficial genera, including Faecalibacterium, Bifidobacterium , and [Eubacterium]_coprostanoligenes_group , whereas Escherichia-Shigella and Bacteroides were enriched in sedentary individuals. Functional prediction revealed a higher abundance of amino acid biosynthesis and metabolism-related pathways in athletes. Among five supervised machine learning algorithms evaluated, the Random Forest model achieved the best classification performance, with an area under the receiver operating characteristic curve of 0.96 in the discovery cohort and 0.86 in the external validation cohort. Lachnospiraceae_NC2004_group was identified as the most important discriminatory feature. Conclusion Long-term exercise training is associated with distinct gut microbial signatures, a healthier microbial ecosystem, and enhanced predicted metabolic functions. Exercise-associated gut microbial signatures accurately distinguish athletes from sedentary adults and may serve as candidate biomarkers for monitoring exercise adaptation and supporting precision exercise interventions.