Microbial marker combination for detecting children obstructive sleep apnea and application thereof

Abstract

The invention relates to the field of biotechnology and health informatics, and particularly discloses a microbial marker combination for detecting obstructive sleep apnea of children and application of the microbial marker combination. The microbial marker combination comprises the following nine microbial genus: gStreptococcus, gSolobacterium, gSchaalia, gCapnococcus, gStomatobacterium, gMogibacterium, gClostridia UCG-014, gSelenobacterium, and gPseudoleptotrichum, and the microbial marker combination comprises the following nine microbial genus: gStreptococcus, gSolobacterium, gSchaalia, gCapnococcus, gStomatobacterium, gMogibacterium, gClostridia UCG-014, gSelenobacterium, and gPseudoleptococcus. By detecting the relative abundance of the microorganisms in the saliva sample of the child, a classification model is constructed by adopting a random forest algorithm, non-invasive and objective screening of the OSA of the child is realized, and the method has relatively high accuracy and clinical application value.

Description

Technical Field

[0001] This invention relates to the fields of biotechnology and health informatics, and more specifically, to a combination of microbial biomarkers and their application in risk assessment and auxiliary diagnosis of obstructive sleep apnea (OSA) in children. Background Technology

[0002] Obstructive sleep apnea (OSA) is a common sleep disorder characterized by recurrent pauses or insufficient ventilation during sleep, and its incidence is increasing in children. Long-term OSA can lead to a range of serious problems in children, including impaired neurocognitive development, cardiovascular complications, and growth retardation.

[0003] Currently, the gold standard for diagnosing childhood obstructive sleep apnea (OSA) is nocturnal polysomnography (PSA). However, PSA is a complex and expensive procedure that requires children to wear multiple sensors overnight in a specialized sleep laboratory, causing significant psychological burden and discomfort, resulting in poor clinical adoption and low patient compliance. Furthermore, existing screening questionnaires (such as the Children's Sleep Habits Questionnaire) rely primarily on parental recollection, limiting their accuracy and specificity.

[0004] In recent years, microbiome research has shown a bidirectional "gut-brain axis" communication between gut microbiota and sleep regulation. However, research on the more easily collected oral saliva microbiota, especially its association with OSA, remains lacking. Saliva sample collection is completely non-invasive, simple, and inexpensive, making it an ideal source of samples for childhood health screening. Currently, existing technologies have not revealed saliva microbial biomarkers that can effectively indicate the risk of OSA in children, and there is a lack of objective, non-invasive assessment methods based on such biomarkers. Summary of the Invention

[0005] To address the aforementioned issues, this invention provides a non-invasive and objective children's sleep health screening tool based on a combination of oral microbial biomarkers, generating a "risk assessment report" that can be applied to health management, early warning, and effect monitoring, demonstrating its practicality.

[0006] The specific technical solution adopted in this invention is as follows:

[0007] In a first aspect, the present invention provides a combination of microbial biomarkers for detecting obstructive sleep apnea in children, which is an oral microbial biomarker combination, specifically including the following nine genera of microorganisms:

[0008] Streptococcus, Solobacterium, Schaalia, Capnocytophaga, Stomatobaculum, Mogibacterium, Clostridia UCG-014, Selenomonas, and Pseudoleptotrichia.

[0009] Secondly, this invention provides the application of the aforementioned microbial biomarker combination in constructing a predictive model for obstructive sleep apnea in children. The predictive model for obstructive sleep apnea in children uses a classification model constructed with a random forest algorithm, based on the relative abundance of the nine microbial genera in the saliva samples of the tested children, to determine whether the tested children suffer from OSA.

[0010] Thirdly, the present invention provides a method for constructing a prediction model for obstructive sleep apnea in children, comprising the following steps:

[0011] S1. Data Collection: Saliva samples were collected from children aged 6-12 years. The training set included samples from the healthy group and the OSA group. The relative abundance of microorganisms at the genus level was detected in the samples.

[0012] S2. Feature matrix construction: The feature matrix X is constructed using the relative abundance of all microbial genera in the sample, and the label vector y is constructed using the group labels diagnosed by polysomnography.

[0013] S3. Train a random forest model containing T decision trees using the feature matrix X and the label y. The key parameters of the random forest model and their typical settings are as follows:

[0014] Number of decision trees: T=500; Number of candidate features for node splitting: M represents the total number of features; maximum depth of the decision tree: unrestricted, until the stopping condition is met; minimum number of split samples per node: 2; random seed: 42;

[0015] S4. Feature selection: The average precision descent method is used to calculate the feature importance score of each microbial genus. The features are sorted in descending order of importance score, and the top N important features are used as input to retrain the random forest model. The AUC value is evaluated on the training set. When N=9, the model AUC reaches its peak, and the nine microbial genera are determined as the final biomarker combination.

[0016] S5. Model Consolidation: Using the identified features of the 9 microbial genera and all training samples, retrain the random forest classifier according to the parameters in step S3 and save the model.

[0017] Furthermore, in step S1, saliva samples are collected from the children after they wake up in the morning, before they brush their teeth, rinse their mouths, eat, and drink water.

[0018] Further, in step S1, the relative abundance is the proportion of the number of sequences from a single microbial genus to the total number of sequences in the sample at the genus level. The detection of this relative abundance is well known to those skilled in the art and can be obtained by extracting genomic DNA from the sample and performing 16S rRNA gene sequencing.

[0019] Fourthly, the present invention provides a childhood obstructive sleep apnea prediction model constructed using the aforementioned construction method.

[0020] Fifthly, the present invention provides a system for predicting obstructive sleep apnea in children, comprising the following modules:

[0021] The data input module is used to input the abundance data of the microbial biomarker combination in the obtained subject biological samples;

[0022] A database storage module for storing abundance data of the microbial biomarker combinations in biological samples of a population, including OSA patients and healthy subjects;

[0023] The disease prediction module is connected to the data input module and the database storage module, respectively, and is used to construct a prediction model using the abundance data of the combination of microbial markers in the biological samples of the population, and to diagnose whether the subject has OSA based on the abundance data of the combination of microbial markers of the subject obtained from the data input module.

[0024] The output module outputs a risk assessment report, which includes the probability that the subject belongs to the OSA group or the healthy group, as well as the final determination of whether the subject belongs to the OSA group or the healthy group.

[0025] In a sixth aspect, the present invention provides a computer-readable storage medium storing a computer program that, when executed by a processor, enables the system for predicting obstructive sleep apnea in children to perform its functions.

[0026] In a seventh aspect, the present invention provides a computer device including a memory and a processor, the memory storing a computer program and the processor executing the computer program to perform the functions of the system for predicting obstructive sleep apnea in children.

[0027] Eighthly, the present invention provides the use of reagents for detecting the abundance of said microbial biomarker combinations in the preparation of a kit for screening obstructive sleep apnea in children.

[0028] Furthermore, the kit for screening obstructive sleep apnea in children may include genomic DNA extraction reagents and reagents for detecting the abundance of the microbial biomarker combination; the reagents for detecting the abundance of the microbial biomarker combination include primers and / or probes capable of amplifying these microbial biomarkers. Genomic DNA extraction reagents and abundance detection reagents are well known to those skilled in the art and can be obtained through conventional means, and therefore are not particularly limited thereto.

[0029] Compared with the prior art, the beneficial effects and significant progress of the present invention are as follows:

[0030] 1. This invention is the first to discover a set of oral salivary microbial biomarkers significantly associated with obstructive sleep apnea in children, including the following nine genera: *Streptococcus*, *Solobacterium*, *Schaalia*, *Capnocytophaga*, *Stomatobaculum*, *Mogibacterium*, *Clostridia_UCG-014*, *Selenomonas*, and *Pseudoleptotrichia*. By detecting the abundance characteristics of this set of microbial biomarkers, the risk of obstructive sleep apnea in children can be effectively predicted. Experiments have demonstrated that the predictive model constructed based on this biomarker set has high accuracy.

[0031] 2. The obstructive sleep apnea-related microbial biomarkers of the present invention are detected based on salivary microbial sequencing data. After rigorous data screening and verification, the results are accurate and safe.

[0032] 3. This invention provides a method for constructing a predictive model for obstructive sleep apnea, which constructs a random forest model with higher specificity, screening efficiency and accuracy based on a set of screened auxiliary diagnostic microbial biomarkers.

[0033] 4. The combination of obstructive sleep apnea-related microbial markers of the present invention can be used to prepare diagnostic reagents or kits for obstructive sleep apnea, for non-invasive auxiliary diagnosis of pediatric OSA patients.

[0034] 5. The present invention has a clear application scenario and a non-invasive sampling method; it is particularly suitable for large-scale preliminary screening and long-term, dynamic biological monitoring of the effects of sleep intervention measures. Attached Figure Description

[0035] Figure 1 Figure 1: Venn diagram of microbial species in the morning saliva of children in the obstructive sleep apnea group and the healthy group in Example 1.

[0036] Figure 2 Heatmap of microbial species in the saliva of children in the obstructive sleep apnea group and the healthy group in Example 1.

[0037] Figure 3 ROC curve of the training set in Example 2.

[0038] Figure 4 ROC curves of the independent validation set in Example 3.

[0039] Figure 5 Figure 4: Screening results of obstructive sleep apnea in a validation sample from Example 4. Detailed Implementation

[0040] The present invention will be further described below with reference to the accompanying drawings and specific embodiments. Unless otherwise specified, the reagents used in the following embodiments are all commercially available conventional reagents, and the experimental procedures involved are all conventional procedures in the art unless otherwise specified.

[0041] Example 1: Preliminary Comparative Analysis of Oral Microbiota Differences between Children with OSA and Healthy Controls

[0042] 1. Data collection

[0043] Inclusion criteria:

[0044] (1) Age 6-12 years old;

[0045] (2) According to the "Guidelines for the Diagnosis and Treatment of Obstructive Sleep Apnea in Children", OSA was diagnosed after overnight polysomnography;

[0046] (3) Has not received OSA-related treatment (such as adenoidectomy, continuous positive airway pressure therapy, etc.).

[0047] Exclusion criteria:

[0048] (1) Use of antibiotics, probiotics or prebiotics within the past 4 weeks;

[0049] (2) Had an acute respiratory infection or oral inflammation within the past two weeks;

[0050] (3) Suffering from congenital heart disease, nervous system disease, hereditary metabolic disease or other serious chronic systemic disease;

[0051] (4) The patient has maxillofacial developmental malformation or a diagnosed genetic syndrome (such as Down syndrome).

[0052] (5) Unable to cooperate in completing saliva sample collection.

[0053] 2. Saliva Sample Collection and Processing

[0054] Number of saliva samples: 72 samples in total, including 34 samples from the healthy group and 38 samples from the OSA group.

[0055] All samples were collected from the children upon waking, before brushing their teeth, rinsing their mouths, eating, and drinking to ensure the original state of their saliva microbiome. Children were instructed to spit their non-irritating saliva into the collection tubes, ensuring a contamination-free process throughout.

[0056] 3. DNA extraction and 16S rRNA gene sequencing

[0057] (1) DNA extraction: After genomic DNA extraction was completed using the EZNA® soil DNA kit (Omega Bio-tek, Norcross, GA, US), the extracted genomic DNA was detected by 1% agarose gel electrophoresis.

[0058] (2) PCR amplification: PCR amplification was performed using universal primers targeting the V3-V4 region of the 16S rRNA gene (upstream primer 338F (5'-ACTCCTACGGGAGGCAGCAG-3'); downstream primer 806R (5'-GGACTACHVGGGTWTCTAAT-3')). The amplification products were quality checked by agarose gel electrophoresis.

[0059] (3) Quantitative fluorescence: Referring to the preliminary quantitative results of electrophoresis, the PCR products were detected and quantified using the QuantiFluor™-ST blue fluorescence quantitative system (Promega). Then, according to the sequencing requirements of each sample, the different samples were mixed in the appropriate proportion to form a library for sequencing (different samples were distinguished by the index on the 5' end of the primer).

[0060] (4) Sequencing: The constructed library was quantified and homogenized and then subjected to PE250 paired-end sequencing.

[0061] 4. Bioinformatics Analysis Workflow

[0062] After the PE reads obtained from sequencing were split, the paired-end reads were first quality-controlled and filtered according to sequencing quality. Simultaneously, the reads were spliced ​​based on their overlap relationship to obtain optimized data after quality control and splicing. Then, sequence denoising methods (DADA2 / Deblur) were used to process the optimized data to obtain representative ASV (Amplicon Sequence Variant) sequences and relative abundance information. Based on the ASV representative sequences and relative abundance information, community diversity analysis and species difference analysis were performed.

[0063] 5. Results

[0064] Venn diagrams can be used to count the number of species that are common to and unique to a sample, and can intuitively show the similarity and overlap of species composition between two groups of samples. Figure 1 The Venn diagram shows that there are 93 common species in the two groups of samples, 15 species unique to the healthy group, and 18 species unique to the OSA group.

[0065] A community heatmap uses color gradients to represent the size of data in a two-dimensional matrix or table and presents information about the species composition of a community. Typically, clustering is performed based on the similarity of abundance between species or samples, and the results are displayed on the community heatmap, allowing high-abundance and low-abundance species to be grouped together. Color variations and similarity levels reflect the similarity and differences in community composition at various taxonomic levels between two groups of samples. Figure 2 The heatmap shown illustrates the distribution of the top dominant species in the OSA group and the healthy group, with inconsistent species trends between the two groups.

[0066] The above results indicate that there are significant differences in oral microbiota between the OSA group and the healthy group, providing a basis for distinguishing children with OSA from healthy individuals based on oral microbiota.

[0067] Example 2: Screening and Model Construction of Oral Microbial Biomarker Combinations

[0068] 1. Optimal Risk Screening Model Construction and Training Methods

[0069] This embodiment uses the random forest algorithm to construct a child OSA risk classification model. The training of the model is a data-driven process based on explicit rules and parameters, and the specific steps are as follows:

[0070] (1) Data preparation and feature definition:

[0071] A feature matrix X was constructed using the relative abundance data of oral microorganisms at the genus level from all training samples (i.e., the 72 samples from Example 1 as the training set). Each row represents a sample, and each column represents a microbial genus (feature). The sample category label y is either "OSA group" or "healthy group" as diagnosed by PSG.

[0072] (2) Training process of random forest model:

[0073] A random forest containing T decision trees is trained using the feature matrix X and the labels y. A single decision tree h k The construction follows the following reproducible rules:

[0074] ① Bootstrap sampling: Randomly sample N samples with replacement from the original training set (N is the total number of samples in the training set) to form a training subset of the tree. This process is reproducible by setting a fixed random seed (random_state).

[0075] ② Node splitting rule: When splitting at each non-leaf node of the tree: First, randomly select m features from all M microbial features as a subset of candidate features. m is a key parameter of the model, typically set as follows:

[0076]

[0077] Then, from these m candidate features, the feature that maximizes the reduction of Gini impurity after splitting and its splitting threshold are selected. The formula for calculating the Gini impurity of node t is:

[0078]

[0079] Where c is the number of categories (C=2 in this invention), and p(i|t) is the proportion of samples belonging to category I in node t.

[0080] ③ Tree growth stopping rules: When the tree reaches the preset maximum depth (max_depth), or the number of samples contained in a node is less than the minimum number of split samples (min_samples_split), or the Gini impurity of a node is 0, splitting stops, and the node is marked as a leaf node. The output of a leaf node is the class of the majority of samples in that node.

[0081] (3) Model parameter setting and fixing:

[0082] In this embodiment, the key parameters of the random forest model and their typical settings are as follows:

[0083] Number of decision trees (n_estimators): T=500;

[0084] Number of candidate features for node splitting (max_features): ;

[0085] Maximum depth of decision tree (max_depth): No limit, until the stopping condition is met;

[0086] Minimum number of split samples per node (min_samples_split): 2;

[0087] Random seed (random_state): 42 (The purpose of fixing the random seed is to ensure that the Bootstrap sampling and node random feature selection processes are completely repeatable).

[0088] (4) Model prediction (voting) mechanism:

[0089] For a new saliva sample X new, the OSA risk assessment process is as follows:

[0090] The relative abundance feature vector of the microbial genera in this sample is input into each decision tree h in the trained random forest. k .

[0091] Each tree h k Produce an independent prediction category c k (OSA group or healthy group).

[0092] Random forests integrate the predictions of all trees using a majority voting method to obtain the final category. :

[0093]

[0094] in, This is an indicator function; its value is 1 when the condition within the parentheses is true, and 0 otherwise. Final output: This represents the category with the most votes, and can also output the probability of belonging to the OSA category. .

[0095] 2. Feature selection and determination of optimal combination

[0096] To obtain a concise and efficient model, we perform feature importance analysis based on the trained random forest model described above.

[0097] (1) Feature importance assessment: The feature importance score for each microbial genus was calculated using the average precision decline method. This score reflects the degree to which the model prediction accuracy decreases after the value of the feature is randomly shuffled in the model prediction. The greater the decrease, the more important the feature.

[0098] (2) Determining the optimal feature combination: We sorted all microbial genera in descending order of importance score, and retrained the random forest model by taking the top N important features as input, and calculated its AUC value on the training set. The results showed that when the number of variables with the top importance ranking N=9, the model performance (AUC) of the random forest reached its peak when using this number of variables. Therefore, the minimum optimal feature combination for constructing the final screening model was determined, namely the following 9 microbial genera:

[0099] g__Streptococcus, g__Solobacterium, g__Schaalia, g__Capnocytophaga, g__Stomatobaculum, g__Mogibacterium, g__Clostridia_UCG-014, g__Selenomonas, g__Pseudoleptotrichia.

[0100] (3) Final model solidification: Using these 9 key features and all training samples, retrain the final random forest classifier according to the process and parameters described in Part 1, and save the model (including tree structure, splitting rules, parameters, etc.) for subsequent clinical sample evaluation.

[0101] The classification accuracy of the combined characteristics of this group, which contains nine microorganisms, was assessed. The results are as follows: Figure 3 As shown, the point marked on the ROC curve represents the optimal critical value (Specificity = 0.73 and Sensitivity = 0.88); the AUC is the area under the corresponding curve: 0.75. This indicates that the classification prediction model has a certain degree of discriminative accuracy.

[0102] Example 3: Validation of OSA Detection Effect of Microbial Markers and Models

[0103] 1. Study subjects and sample collection: Saliva samples were collected from 20 children, including 10 OSA patients and 10 healthy individuals.

[0104] 2. Sample collection and sequencing: As before (steps 3-4 in Example 1), obtain 16S rRNA gene sequencing data from saliva samples and calculate their relative abundance at the genus level.

[0105] Feature extraction: From the above results, the relative abundance values ​​of the nine key microbial genera are extracted to form a nine-dimensional feature vector.

[0106] 3. Diagnostic efficacy analysis: Receiver operating curves were plotted to analyze the sensitivity and specificity of the selected microbial biomarker combinations in the validation set between the healthy group and OSA subjects, in order to determine their diagnostic efficacy for childhood OSA.

[0107] Using the abundance combination of microbial biomarkers as the detection variable, ROC curves were plotted and their AUC values ​​were calculated. The results are as follows: Figure 4 As shown, the AUC value was 0.710, the specificity was 0.75, and the sensitivity was 0.786. This indicates that the combination of salivary microbial markers can serve as microbial markers for diagnosing OSA in children.

[0108] Example 4: Diagnostic Case of Borderline Difficult Samples

[0109] The OSA risk assessment process for a child to be screened is as follows:

[0110] 1. Sample collection and sequencing: As before, obtain 16S rRNA gene sequencing data from saliva samples and calculate their relative abundance at the genus level.

[0111] 2. Feature extraction: From the above results, the relative abundance values ​​of the nine key microbial genera are extracted to form a nine-dimensional feature vector.

[0112] 3. Model prediction: Input the 9-dimensional feature vector into the final random forest model that has been fixed and saved in step (3) of Example 2 (the model file contains the structure, split points and parameters of all decision trees).

[0113] 4. Output Results: The model runs all its internal decision trees and performs majority voting, which can generate results such as... Figure 5 The visualization report shown displays the prediction results and the probability of belonging to each category. The group with the highest probability is taken as the grouping result in the final report. In this specific embodiment, the final output is that the child belongs to the "OSA group" and the predicted probability of belonging to this group - "Prob_OSA".

[0114] Figure 5 The predicted sample results table shows that the predicted sample belongs to the OSA group. The gold standard for diagnosis, PSG, confirmed that the child did indeed have chronic intermittent hypoxia and sleep fragmentation during nighttime sleep, consistent with the diagnosis of obstructive sleep apnea. Moreover, the child was near the diagnostic borderline, making assessment difficult. The predicted results of this method for this child are consistent with the actual situation, indicating that the OSA screening microbial biomarker combination and model of this invention have high reliability.

[0115] This specific embodiment is merely an explanation of the present invention and is not intended to limit the present invention. Any changes made by those skilled in the art after reading the specification of the present invention, as long as they are within the scope of the claims of the present invention, will be protected by patent law.

Claims

1. A combination of microbial biomarkers for detecting obstructive sleep apnea in children, characterized in that, It is a combination of oral microbial markers, including the following 9 genera: Streptococcus, Sorafenib, Salmonella, Carbonylphage, Oral Bacillus, Diplobacterium, Clostridium UCG-014, Lunatomium, and Pseudomonas.

2. The application of the microbial biomarker combination according to claim 1 in constructing a predictive model for obstructive sleep apnea in children, characterized in that, The childhood obstructive sleep apnea prediction model uses a classification model constructed with a random forest algorithm based on the relative abundance of the nine microbial genera in the saliva samples of the tested children to determine whether the tested children have OSA.

3. A method for constructing a predictive model for obstructive sleep apnea in children, characterized in that, Includes the following steps: S1. Data Collection: Saliva samples were collected from children aged 6-12 years to detect the relative abundance of horizontal microorganisms in the samples. The training set included samples from the healthy group and the OSA group. S2. Feature matrix construction: The feature matrix X is constructed using the relative abundance of all microbial genera in the sample, and the label vector y is constructed using the group labels diagnosed by polysomnography. S3. Train a random forest model containing T decision trees using the feature matrix X and the label y. The key parameters of the random forest model and their typical settings are as follows: Number of decision trees: T=500; Number of candidate features for node splitting: M represents the total number of features; maximum depth of the decision tree: unlimited, until the stopping condition is met; minimum number of split samples per node: 2; random seed: 42; S4. Feature selection: The feature importance score of each microbial genus is calculated using the average precision descent method, and the features are sorted in descending order of importance score. The top N important features are then used as input to retrain the random forest model. The AUC value is evaluated on the training set. When N=9, the model AUC reaches its peak, and the nine microbial genera described in claim 1 are determined as the final biomarker combination. S5. Model Consolidation: Using the identified features of the 9 microbial genera and all training samples, retrain the random forest classifier according to the parameters in step S3 and save the model.

4. The construction method according to claim 3, characterized in that, In step S1, saliva samples were collected from the children after they woke up in the morning, before they brushed their teeth, rinsed their mouths, ate, and drank water; genomic DNA was extracted from the samples and 16S rRNA sequencing was performed to obtain relative abundance data of microorganisms at the genus level.

5. A childhood obstructive sleep apnea prediction model constructed using the construction method described in claim 3.

6. A system for predicting obstructive sleep apnea in children, characterized in that, Includes the following modules: The data input module is used to input the abundance data of the combination of microbial markers as described in claim 1 in the obtained biological samples of the subjects; A database storage module for storing abundance data of the microbial biomarker combinations in biological samples of a population, including OSA patients and healthy subjects; The disease prediction module is connected to the data input module and the database storage module, respectively, and is used to construct a prediction model using the abundance data of the combination of microbial markers in the biological samples of the population, and to diagnose whether the subject has OSA based on the abundance data of the combination of microbial markers of the subject obtained from the data input module. The output module outputs a risk assessment report, which includes the probability that the subject belongs to the OSA group or the healthy group, as well as the final determination of whether the subject belongs to the OSA group or the healthy group.

7. A computer-readable storage medium, characterized in that, It stores a computer program that, when executed by a processor, can perform the functions of the system as described in claim 6.

8. A computer device, characterized in that, It includes a memory and a processor, the memory storing a computer program and the processor executing the computer program to perform the functions of the system as described in claim 6.

9. The use of the reagent for detecting the abundance of the microbial biomarker combination of claim 1 in the preparation of a kit for screening obstructive sleep apnea in children.

Images