Meta‐analysis of caries microbiome studies can improve upon disease prediction outcomes

Abstract

As one of the most prevalent infective diseases worldwide, it is crucial that we not only know the constituents of the oral microbiome in dental caries but also understand its functionality. Herein, we present a reproducible meta‐analysis to effectively report the key components and the associated functional signature of the oral microbiome in dental caries. Publicly available sequencing data were downloaded from online repositories and subjected to a standardized analysis pipeline before analysis. Meta‐analyses identified significant differences in alpha and beta diversities of carious microbiomes when compared to healthy ones. Additionally, machine learning and receiver operator characteristic analysis showed an ability to discriminate between healthy and disease microbiomes. We identified from importance values, as derived from random forest analyses, a group of genera, notably containing Selenomonas, Aggregatibacter, Actinomyces and Treponema, which can be predictive of dental caries. Finally, we propose the most appropriate study design for investigating the microbiome of dental caries by synthesizing the studies, which had the most accurate differentiation based on random forest modelling. In conclusion, we have developed a non‐biased, reproducible pipeline, which can be applied to microbiome meta‐analyses of multiple diseases, but importantly we have derived from our meta‐analysis a key group of organisms that can be used to identify individuals at risk of developing dental caries based on oral microbiome inhabitants.

Limitations of using 16S rRNA microbiome sequencing to predict oral squamous cell carcinoma

“…These then underwent a two‐step process for verification and inclusion by another laboratory‐based clinician (SS). Shortlisted articles were then assigned a score from 1 to 5 based on data access and cohort metadata inclusion as outlined previously (19).…”

“…Diversity and compositional analyses were carried out as previously described (19). Briefly, α‐ and β‐diversity analyses were calculated using phyloseq.…”

“…1. A full methodology has been included as supplementary material and significant deviations have been briefly listed here (19). Studies were collected by using keyword searches on PubMed and Web of Science (Clarivate Analytics, Philadelphia, Pensylvania, USA) to select for microbiome and NGS studies, performed on the oral cavity, and specifically related to oral squamous cell carcinoma (Table S1).…”

“…As described previously, four high-impact case-controlled studies were selected as training datasets for random forest classifiers (19,26). The data was randomly split into a 80:20 ratio to create a training and validation dataset.…”

There Is More to Wounds than Bacteria: Fungal Biofilms in Chronic Wounds

“…PubMed search in March 2022 (data not shown) using the following terms adapted from a recent microbiome metaanalysis [6]: ((wound OR diabet* foot OR diabet* foot ulcer OR DFU OR laceration OR ischemic OR neuropathic OR pressure) NOT (review)) AND ((microb* OR bacteri* OR archea* OR fung* OR mycob*) AND (structure OR composition OR diversity OR community) AND (sequencing OR metabarcoding OR amplicon OR metagenom* OR 16S OR "ITS")). In a large-scale microbiome study performed by Wolcott and colleagues (2016), the microbiome of 2963 chronic wound samples from DFUs, nonhealing surgical wounds, venous leg ulcers, and decubitus ulcers was defined [7].…”