Multi-omics analysis of associations between host demographics and saliva metabolome, sugar profiles, and microbiome profiles

Abstract

Abstract Omics profiling of saliva is an emerging research area with potential to uncover molecular signatures associated with oral and systemic health. We conducted a comprehensive multi-omics analysis of saliva to investigate associations between host demographics (age, sex, body mass index (BMI)) and molecular profiles. Saliva from 423 participants (16–79-years-old) were analyzed using LC-MS metabolomics (9,380 metabolite features for 416 participants), GC×GC-MS sugar profiling (69 sugars for 200 participants), and full-length 16S rDNA sequencing (500 microbial species for 420 participants). We used random forest modeling, multivariate OPLS analysis, and partial correlation networks for data integration. Age emerged as the strongest demographic factor, explaining up to 30% of variance in metabolite features, 17% in sugars, and 25% in microbial species, while sex showed moderate and BMI minimal associations. Age-associated metabolites included caffeine and trigonelline (higher in older participants) and urocanic acid (higher in younger participants). Younger participants had greater abundance of saccharolytic, facultative anaerobic bacteria while older participants had more anaerobic species. Species in the Streptococcus , Prevotella , and Veillonella genera correlated strongly with salivary sugars. These findings demonstrate that saliva provides a rich source of molecular information related to the individual, and that demographic factors must be considered in saliva-based biomarker-discovery studies.