skip to main content

Title: Machine Learning Approach Identified Multi-Platform Factors for Caries Prediction in Child-Mother Dyads
Untreated tooth decays affect nearly one third of the world and is the most prevalent disease burden among children. The disease progression of tooth decay is multifactorial and involves a prolonged decrease in pH, resulting in the demineralization of tooth surfaces. Bacterial species that are capable of fermenting carbohydrates contribute to the demineralization process by the production of organic acids. The combined use of machine learning and 16s rRNA sequencing offers the potential to predict tooth decay by identifying the bacterial community that is present in an individual’s oral cavity. A few recent studies have demonstrated machine learning predictive modeling using 16s rRNA sequencing of oral samples, but they lack consideration of the multifactorial nature of tooth decay, as well as the role of fungal species within their models. Here, the oral microbiome of mother–child dyads (both healthy and caries-active) was used in combination with demographic–environmental factors and relevant fungal information to create a multifactorial machine learning model based on the LASSO-penalized logistic regression. For the children, not only were several bacterial species found to be caries-associated ( Prevotella histicola, Streptococcus mutans , and Rothia muciloginosa ) but also Candida detection and lower toothbrushing frequency were also caries-associated. Mothers enrolled more » in this study had a higher detection of S. mutans and Candida and a higher plaque index. This proof-of-concept study demonstrates the significant impact machine learning could have in prevention and diagnostic advancements for tooth decay, as well as the importance of considering fungal and demographic–environmental factors. « less
; ; ; ; ; ; ;
Award ID(s):
Publication Date:
Journal Name:
Frontiers in Cellular and Infection Microbiology
Sponsoring Org:
National Science Foundation
More Like this
  1. Early childhood caries (ECC) is not only the most common chronic childhood disease but also disproportionately affects underserved populations. Of those, children living in Thailand have been found to have high rates of ECC and severe ECC. Frequently, the cause of ECC is blamed on a handful of cariogenic organisms, such as Streptococcus mutans and Streptococcus sobrinus . However, ECC is a multifactorial disease that results from an ecological shift in the oral cavity from a neutral pH (~7.5) to an acidic pH (<5.5) environment influenced by the host individual’s biological, socio-behavioral, and lifestyle factors. Currently, there is a lack of understanding of how risk factors at various levels influence the oral health of children at risk. We applied a statistical machine learning approach for multimodal data integration (parallel and hierarchical) to identify caries-related multiplatform factors in a large cohort of mother-child dyads living in Chiang Mai, Thailand (N=177). Whole saliva (1 mL) was collected from each individual for DNA extraction and 16S rRNA sequencing. A set of maternal and early childhood factors were included in the data analysis. Significantly, vaginal delivery, preterm birth, and frequent sugary snacking were found to increase the risk for ECC. The salivary microbial diversitymore »was significantly different in children with ECC or without ECC. Results of linear discriminant analysis effect size (LEfSe) analysis of the microbial community demonstrated that S. mutans , Prevotella histicola , and Leptotrichia hongkongensis were significantly enriched in ECC children. Whereas Fusobacterium periodonticum was less abundant among caries-free children, suggesting its potential to be a candidate biomarker for good oral health. Based on the multimodal data integration and statistical machine learning models, the study revealed that the mode of delivery and snack consumption outrank salivary microbiome in predicting ECC in Thai children. The biological and behavioral factors may play significant roles in the microbial pathobiology of ECC and warrant further investigation.« less
  2. Despite the cariogenic role of Candida suggested from recent studies, oral Candida acquisition in children at high risk for early childhood caries (ECC) and its association with cariogenic bacteria Streptococcus mutans remain unclear. Although ECC disproportionately afflicts socioeconomically disadvantaged and racial-minority children, microbiological studies focusing on the underserved group are scarce. Our prospective cohort study examined the oral colonization of Candida and S. mutans among 101 infants exclusively from a low-income and racial-minority background in the first year of life. The Cox hazard proportional model was fitted to assess factors associated with the time to event of the emergence of oral Candida and S. mutans. Oral Candida colonization started as early as 1 wk among 13% of infants, increased to 40% by 2 mo, escalated to 48% by 6 mo, and remained the same level until 12 mo. S. mutans in saliva was detected among 20% infants by 12 mo. The emergence of S. mutans by year 1 was 3.5 times higher (hazard ratio [HR], 3.5; confidence interval [CI], 1.1–11.3) in infants who had early colonization of oral Candida compared to those who were free of oral Candida ( P = 0.04) and 3 times higher (HR, 3.0; CI, 1.3–6.9)more »among infants whose mother had more than 3 decayed teeth ( P = 0.01), even after adjusting demographics, feeding, mother’s education, and employment status. Infants’ salivary S. mutans abundance was positively correlated with infants’ Candida albicans ( P < 0.01) and Candida krusei levels ( P < 0.05). Infants’ oral colonization of C. albicans was positively associated with mother’s oral C. albicans carriage and education ( P < 0.01) but negatively associated with mother’s employment status ( P = 0.01). Future studies are warranted to examine whether oral Candida modulates the oral bacterial community as a whole to become cariogenic during the onset and progression of ECC, which could lead to developing novel ECC predictive and preventive strategies from a fungal perspective.« less
  3. Other
    Bacterial biofilms associated with implants remain a significant source of infections in dental, implant, and other healthcare industries due to challenges in biofilm removal. Biofilms consist of bacterial cells surrounded by a matrix of extracellular polymeric substance (EPS), which protects the colony from many countermeasures, including antibiotic treatments. Biofilm EPS composition is also affected by environmental factors. In the oral cavity, the presence of sucrose affects the growth of Streptococcus mutans that produce acids, eroding enamel and forming dental caries. Biofilm formation on dental implants commonly leads to severe infections and failure of the implant. This work determines the effect of sucrose concentration on biofilm EPS formation and adhesion of Streptococcus mutans, a common oral colonizer. Bacterial biofilms are grown with varying concentrations of sucrose on titanium substrates simulating dental implant material. Strategies for measuring adhesion for films such as peel tests are inadequate for biofilms, which have low cohesive strength and will fall apart when tensile loading is applied directly. The laser spallation technique is used to apply a stress wave loading to the biofilm, causing the biofilm to delaminate at a critical tensile stress threshold. Biofilm formation and EPS structures are visualized at high magnification with scanningMore>>

    Host-associated microbial communities can influence physiological processes of macroorganisms, including contributing to infectious disease resistance. For instance, some bacteria that live on amphibian skin produce antifungal compounds that inhibit two lethal fungal pathogens, Batrachochytrium dendrobatidis (Bd) and Batrachochytrium salamandrivorans (Bsal). Therefore, differences in microbiome composition among host species or populations within a species can contribute to variation in susceptibility to Bd/Bsal. This study applies 16S rRNA sequencing to characterize the skin bacterial microbiomes of three widespread terrestrial salamander genera native to the western United States. Using a metacommunity structure analysis, we identified dispersal barriers for these influential bacteria between salamander families and localities. We also analysed the effects of habitat characteristics such as percent natural cover and temperature seasonality on the microbiome. We found that certain environmental variables may influence the skin microbial communities of some salamander genera more strongly than others. Each salamander family had a somewhat distinct community of putative anti-Bd skin bacteria, suggesting that salamanders may select for a functional assembly of cutaneous symbionts that could differ in its ability to protect these amphibians from disease. Our observations raise the need to consider host identity and environmental heterogeneity during the selection of probiotics to treat wildlifemore »diseases.

    « less
  5. Effective coral restoration must include comprehensive investigations of the targeted coral community that consider all aspects of the coral holobiont—the coral host, symbiotic algae, and microbiome. For example, the richness and composition of microorganisms associated with corals may be indicative of the corals’ health status and thus help guide restoration activities. Potential differences in microbiomes of restoration corals due to differences in host genetics, environmental condition, or geographic location, may then influence outplant success. The objective of the present study was to characterize and compare the microbiomes of apparently healthy Acropora cervicornis genotypes that were originally collected from environmentally distinct regions of Florida’s Coral Reef and sampled after residing within Mote Marine Laboratory’s in situ nursery near Looe Key, FL (USA) for multiple years. By using 16S rRNA high-throughput sequencing, we described the microbial communities of 74 A. cervicornis genotypes originating from the Lower Florida Keys ( n  = 40 genotypes), the Middle Florida Keys ( n  = 15 genotypes), and the Upper Florida Keys ( n  = 19 genotypes). Our findings demonstrated that the bacterial communities of A. cervicornis originating from the Lower Keys were significantly different from the bacterial communities of those originating from the Upper and Middle Keys even aftermore »these corals were held within the same common garden nursery for an average of 3.4 years. However, the bacterial communities of corals originating in the Upper Keys were not significantly different from those in the Middle Keys. The majority of the genotypes, regardless of collection region, were dominated by Alphaproteobacteria, namely an obligate intracellular parasite of the genus Ca. Aquarickettsia . Genotypes from the Upper and Middle Keys also had high relative abundances of Spirochaeta bacteria. Several genotypes originating from both the Lower and Upper Keys had lower abundances of Aquarickettsia , resulting in significantly higher species richness and diversity. Low abundance of Aquarickettsia has been previously identified as a signature of disease resistance. While the low- Aquarickettsia corals from both the Upper and Lower Keys had high abundances of an unclassified Proteobacteria, the genotypes in the Upper Keys were also dominated by Spirochaeta . The results of this study suggest that the abundance of Aquarickettsia and Spirochaeta may play an important role in distinguishing bacterial communities among A. cervicornis populations and compositional differences of these bacterial communities may be driven by regional processes that are influenced by both the environmental history and genetic relatedness of the host. Additionally, the high microbial diversity of low- Aquarickettsia genotypes may provide resilience to their hosts, and these genotypes may be a potential resource for restoration practices and management.« less