Recent advances in genomic technologies have enabled more in-depth study of the oral microbiome. In this study, we compared the amplicons generated by primers targeting different sites of the 16S rRNA gene found in the Human Oral Microbiome Database (HOMD). Six sets of primer targeting V1–V2, V1–V3, V3–V4, V4–V5, V5–V7 and V6–V8 regions of 16S rRNA were tested via in silico simulation. Primers targeting the V1–V2, V3–V4, and V4–V5 regions generated more than 90% of the original input sequences. Primers targeting the V1–V2 and V1–V3 regions exhibited a low number of mismatches and unclassified sequences at the taxonomic level, but there were notable discrepancies at the species level. Phylogenetic tree comparisons showed primers targeting the V1–V2 and V3–V4 regions showed performances similar to primers targeting the whole 16s RNA region in terms of separating total oral microbiomes and periodontopathogens. In an analysis of clinical oral samples, V1–V2 primers showed superior performance for identifying more taxa and had better resolution sensitivity for Streptococcus than V3–V4 primers. In conclusion, primers targeting the V1–V2 region of 16S rRNA showed the best performance for oral microbiome studies. In addition, the study demonstrates the need for careful PCR primer selections.

Several studies have demonstrated that nuclear magnetic resonance (NMR) metabolic profiles can differentiate patients with caries from healthy individuals; however, these studies only identified individual metabolites. The present study aimed to identify a salivary metabolite biomarker panel for the diagnosis of early childhood caries (ECC). Saliva samples from children with and without caries were analyzed using NMR spectroscopy. Multivariate and univariate analyses were performed to identify the discriminating metabolites. Selected metabolites were further evaluated and used to detect ECC. The saliva samples of children with ECC were characterized based on the increased levels of formate, glycerophosphocholine, and lactate and reduced levels of alanine, glycine, isoleucine, lysine, proline, and tyrosine. The levels of these metabolites were significantly different from those in the control in the ECC subgroup according to caries severity and correlated with the number of decayed and filled teeth or surfaces. Subsequently, an optimal salivary metabolite biomarker panel comprising formate, lactate, proline, and glycine was developed. This panel exhibited a better diagnostic performance for ECC than a single metabolite. These results demonstrate that salivary metabolic signatures can reflect oral conditions associated with dental caries, thereby emphasizing the importance of distinct salivary metabolic profiles as potential biomarkers of ECC.