Preliminary analysis of salivary microbiota in catathrenia (nocturnal groaning) using machine learning algorithms

Objectives The present study aimed to characterize the salivary microbiota in patients with catathrenia and to longitudinally validate potential biomarkers after treatment with mandibular advancement devices (MAD).Materials and methods Twenty-two patients with catathrenia (12 M/10 F, median age 28 y...

Full description

Saved in:
Bibliographic Details
Main Authors: Min Yu, Yujia Lu, Wanxin Zhang, Xu Gong, Zeliang Hao, Liyue Xu, Yongfei Wen, Xiaosong Dong, Fang Han, Xuemei Gao
Format: Article
Language:English
Published: Taylor & Francis Group 2025-12-01
Series:Journal of Oral Microbiology
Subjects:
Online Access:https://www.tandfonline.com/doi/10.1080/20002297.2025.2489613
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849725939591675904
author Min Yu
Yujia Lu
Wanxin Zhang
Xu Gong
Zeliang Hao
Liyue Xu
Yongfei Wen
Xiaosong Dong
Fang Han
Xuemei Gao
author_facet Min Yu
Yujia Lu
Wanxin Zhang
Xu Gong
Zeliang Hao
Liyue Xu
Yongfei Wen
Xiaosong Dong
Fang Han
Xuemei Gao
author_sort Min Yu
collection DOAJ
description Objectives The present study aimed to characterize the salivary microbiota in patients with catathrenia and to longitudinally validate potential biomarkers after treatment with mandibular advancement devices (MAD).Materials and methods Twenty-two patients with catathrenia (12 M/10 F, median age 28 y) and 22 age-matched control volunteers (8 M/14 F, median age 30 y) were included in the cross-sectional study. Video/audio polysomnography was conducted for diagnosis. All patients received treatment with custom-fit MAD and were followed for one month. Ten patients (6 M/4 F) underwent post-treatment PSG. Salivary samples were collected, and microbial characteristics were analyzed using 16S rRNA gene sequencing. The 10-fold cross-validated XGBoost and nested Random Forest Classifier machine learning algorithms were utilized to identify potential biomarkers.Results In the cross-sectional study, patients with catathrenia had lower α-diversity represented by Chao 1, Faith’s phylogenetic diversity (pd), and observed species. Beta-diversity based on the Bray-Curtis dissimilarities revealed a significant inter-group separation (p = 0.001). The inter-group microbiota distribution was significantly different on the phylum and family levels. The treatment of MAD did not alter salivary microbiota distribution significantly. Among the most important genera in catathrenia and control classification identified by machine learning algorithms, four genera, Alloprevotella, Peptostreptococcaceae_XI_G1, Actinomyces and Rothia, changed significantly with MAD treatment. Correlation analysis revealed that Alloprevotella was negatively related to the severity of catathrenia (r2= −0.63, p < 0.001).Conclusions High-throughput sequencing revealed that the salivary microbiota composition was significantly altered in patients with catathrenia. Some characteristic genera (Alloprevotella, Peptostreptococcaceae_XI_G1, Actinomyces, and Rothia) could be potential biomarkers sensitive to treatment. Future studies are needed to confirm and determine the mechanisms underlying these findings.
format Article
id doaj-art-9123161e30f14344bf23ee603b8bf8a6
institution DOAJ
issn 2000-2297
language English
publishDate 2025-12-01
publisher Taylor & Francis Group
record_format Article
series Journal of Oral Microbiology
spelling doaj-art-9123161e30f14344bf23ee603b8bf8a62025-08-20T03:10:21ZengTaylor & Francis GroupJournal of Oral Microbiology2000-22972025-12-0117110.1080/20002297.2025.2489613Preliminary analysis of salivary microbiota in catathrenia (nocturnal groaning) using machine learning algorithmsMin Yu0Yujia Lu1Wanxin Zhang2Xu Gong3Zeliang Hao4Liyue Xu5Yongfei Wen6Xiaosong Dong7Fang Han8Xuemei Gao9Department of Orthodontics, Peking University School and Hospital of Stomatology, Beijing, P.R. ChinaDepartment of Orthodontics, Peking University School and Hospital of Stomatology, Beijing, P.R. ChinaDepartment of Orthodontics, Peking University School and Hospital of Stomatology, Beijing, P.R. ChinaDepartment of Orthodontics, Peking University School and Hospital of Stomatology, Beijing, P.R. ChinaDepartment of Stomatology, Xuanwu Hospital, Capital Medical University, Beijing, P.R. ChinaSleep Division, Peking University People’s Hospital, Beijing, P.R. ChinaSleep Division, Peking University People’s Hospital, Beijing, P.R. ChinaSleep Division, Peking University People’s Hospital, Beijing, P.R. ChinaSleep Division, Peking University People’s Hospital, Beijing, P.R. ChinaDepartment of Orthodontics, Peking University School and Hospital of Stomatology, Beijing, P.R. ChinaObjectives The present study aimed to characterize the salivary microbiota in patients with catathrenia and to longitudinally validate potential biomarkers after treatment with mandibular advancement devices (MAD).Materials and methods Twenty-two patients with catathrenia (12 M/10 F, median age 28 y) and 22 age-matched control volunteers (8 M/14 F, median age 30 y) were included in the cross-sectional study. Video/audio polysomnography was conducted for diagnosis. All patients received treatment with custom-fit MAD and were followed for one month. Ten patients (6 M/4 F) underwent post-treatment PSG. Salivary samples were collected, and microbial characteristics were analyzed using 16S rRNA gene sequencing. The 10-fold cross-validated XGBoost and nested Random Forest Classifier machine learning algorithms were utilized to identify potential biomarkers.Results In the cross-sectional study, patients with catathrenia had lower α-diversity represented by Chao 1, Faith’s phylogenetic diversity (pd), and observed species. Beta-diversity based on the Bray-Curtis dissimilarities revealed a significant inter-group separation (p = 0.001). The inter-group microbiota distribution was significantly different on the phylum and family levels. The treatment of MAD did not alter salivary microbiota distribution significantly. Among the most important genera in catathrenia and control classification identified by machine learning algorithms, four genera, Alloprevotella, Peptostreptococcaceae_XI_G1, Actinomyces and Rothia, changed significantly with MAD treatment. Correlation analysis revealed that Alloprevotella was negatively related to the severity of catathrenia (r2= −0.63, p < 0.001).Conclusions High-throughput sequencing revealed that the salivary microbiota composition was significantly altered in patients with catathrenia. Some characteristic genera (Alloprevotella, Peptostreptococcaceae_XI_G1, Actinomyces, and Rothia) could be potential biomarkers sensitive to treatment. Future studies are needed to confirm and determine the mechanisms underlying these findings.https://www.tandfonline.com/doi/10.1080/20002297.2025.2489613Sleep-disordered breathingbiomarkermicrobiotaupper airwayNocturnal groaning
spellingShingle Min Yu
Yujia Lu
Wanxin Zhang
Xu Gong
Zeliang Hao
Liyue Xu
Yongfei Wen
Xiaosong Dong
Fang Han
Xuemei Gao
Preliminary analysis of salivary microbiota in catathrenia (nocturnal groaning) using machine learning algorithms
Journal of Oral Microbiology
Sleep-disordered breathing
biomarker
microbiota
upper airway
Nocturnal groaning
title Preliminary analysis of salivary microbiota in catathrenia (nocturnal groaning) using machine learning algorithms
title_full Preliminary analysis of salivary microbiota in catathrenia (nocturnal groaning) using machine learning algorithms
title_fullStr Preliminary analysis of salivary microbiota in catathrenia (nocturnal groaning) using machine learning algorithms
title_full_unstemmed Preliminary analysis of salivary microbiota in catathrenia (nocturnal groaning) using machine learning algorithms
title_short Preliminary analysis of salivary microbiota in catathrenia (nocturnal groaning) using machine learning algorithms
title_sort preliminary analysis of salivary microbiota in catathrenia nocturnal groaning using machine learning algorithms
topic Sleep-disordered breathing
biomarker
microbiota
upper airway
Nocturnal groaning
url https://www.tandfonline.com/doi/10.1080/20002297.2025.2489613
work_keys_str_mv AT minyu preliminaryanalysisofsalivarymicrobiotaincatathrenianocturnalgroaningusingmachinelearningalgorithms
AT yujialu preliminaryanalysisofsalivarymicrobiotaincatathrenianocturnalgroaningusingmachinelearningalgorithms
AT wanxinzhang preliminaryanalysisofsalivarymicrobiotaincatathrenianocturnalgroaningusingmachinelearningalgorithms
AT xugong preliminaryanalysisofsalivarymicrobiotaincatathrenianocturnalgroaningusingmachinelearningalgorithms
AT zelianghao preliminaryanalysisofsalivarymicrobiotaincatathrenianocturnalgroaningusingmachinelearningalgorithms
AT liyuexu preliminaryanalysisofsalivarymicrobiotaincatathrenianocturnalgroaningusingmachinelearningalgorithms
AT yongfeiwen preliminaryanalysisofsalivarymicrobiotaincatathrenianocturnalgroaningusingmachinelearningalgorithms
AT xiaosongdong preliminaryanalysisofsalivarymicrobiotaincatathrenianocturnalgroaningusingmachinelearningalgorithms
AT fanghan preliminaryanalysisofsalivarymicrobiotaincatathrenianocturnalgroaningusingmachinelearningalgorithms
AT xuemeigao preliminaryanalysisofsalivarymicrobiotaincatathrenianocturnalgroaningusingmachinelearningalgorithms