Comparative Machine Learning Analysis of Saliva and Plaque Microbiomes in Children with Type 1 Diabetes
Discuss this preprint
Start a discussionListed in
This article is not in any list yet, why not save it to one of your lists.Abstract
Background
Type 1 diabetes (T1D) is associated with microbial dysbiosis. While most research has focused on the gut microbiome, limited data address the role of the oral microbiome in T1D. The oral and gut microbiomes overlap substantially, and the oral cavity may influence the gut microbial composition. Saliva and dental plaque represent two distinct oral niches with unique microbial communities, but it remains unclear which better reflects systemic disease states such as T1D. This study compared the performance of salivary and plaque microbiomes in classifying pediatric T1D status.
Methods
Paired saliva and plaque samples were collected from 46 children (23 with T1D and 23 healthy controls). Microbial DNA was extracted and sequenced to target the 16S rRNA gene. The data were processed via QIIME 2 for taxonomic classification and centered log-ratio transformation. Alpha diversity, microbial abundance, and clustering analyses were performed to compare the oral microbiome between the T1D and control groups. Random forest classifiers were used to evaluate and compare the predictive accuracy of saliva- and plaque-based models, both with and without clinical metadata.
Results
Saliva samples presented lower alpha diversity than plaque samples did but presented significantly greater bacterial loads and total microbial abundances. Saliva-based models outperformed plaque-based models, achieving a classification accuracy of 94.2% with or without clinical metadata, compared with 73.3% accuracy for plaque-based models. ROC curve analysis further supported this difference, with saliva models reaching an AUC of approximately 0.94 versus 0.75 for plaque, indicating superior discriminative performance. UMAP clustering revealed more distinct separation of the T1D and control groups in terms of the salivary profiles than in the plaque profiles. Feature importance analysis revealed both unique and shared taxa predictive of T1D in each niche. The incorporation of clinical and demographic metadata did not enhance model performance, underscoring the robustness and predictive strength of microbiome data alone.
Conclusion
The salivary microbiome is a more effective biospecimen than dental plaque for detecting T1D-associated microbial profiles in children. It offers superior classification accuracy and greater sensitivity in distinguishing T1D status, supporting saliva’s potential as a noninvasive, scalable medium for future microbiome-based monitoring.
