A marker combination, combined reagent, and kit for early caries risk assessment in young children, and their application in constructing an early caries risk assessment model for young children

Abstract

The present invention provides a marker combination, a combination reagent, a kit and its application in constructing an early risk assessment model for early caries in young children, belonging to the field of oral preventive medicine and oral public health technology. The present invention uses proteomics and metabolomics to study saliva samples of young children, and finds that 2-oxoadipate, enoyl-CoA hydratase and kynureninase can be used as a marker combination to assess the early risk of caries in young children. The present invention constructs an early risk assessment model for caries in young children based on the abundance reference indexes of 2-oxoadipate, enoyl-CoA hydratase and kynureninase. The early risk assessment model for caries in young children constructed by the method of the present invention has good stability, sensitivity and specificity, and lays a solid foundation for taking effective intervention measures to achieve comprehensive prevention and control of caries.

Description

Technical Field

[0001] The present invention belongs to the field of oral preventive medicine and oral public health technology, and in particular relates to a marker combination, a combination reagent, a kit for early caries risk assessment in young children, and their application in constructing an early caries risk assessment model for young children. Background Art

[0002] Caries in young children is a common and frequently occurring oral disease in children and has become one of the most common childhood health issues worldwide. The results of the 2015 National Oral Health Epidemiological Survey showed that the caries prevalence in children aged 3, 4, and 5 reached 50.8%, 63.6%, and 71.9%, respectively, which is relatively high and has increased significantly compared to 10 years ago. Untreated caries can cause abnormalities in the development and eruption of permanent teeth, affecting children's chewing and pronunciation functions, and impairing their eating and language skills, leading to a series of social and psychological problems and placing a significant economic burden on their families and society. The harm that caries in young children can cause to children's growth, development, and physical and mental health cannot be ignored. According to the four-factor theory of dental caries, dental caries involve the interaction of multiple factors. Existing research has largely focused on single-omics or single-factor analyses. For example, the Cariostat system assesses dental caries risk solely through pH, ​​or constructs dental caries risk assessment models based solely on proteomics and metabolomics. These studies fail to integrate multi-omics techniques to correlate the dynamic changes in dental caries in young children with diverse biomarkers. Consequently, these studies can only reveal biomolecular processes at a specific level, providing limited biological information and failing to comprehensively and comprehensively reflect the interaction mechanisms between the host and microbes. Consequently, the models' efficacy is low. Furthermore, existing diagnostic models are often based on biomarker screening from cross-sectional studies, lacking long-term longitudinal follow-up cohort studies. Therefore, models based on cross-sectional findings are difficult to dynamically monitor dental caries status and assess risk. The existing model, developed during its construction, compares healthy individuals with young children with caries. Therefore, it only analyzes the state after caries develop, diagnosing whether a child has already developed caries by analyzing their saliva. However, it cannot assess and prevent caries in young children before it develops, resulting in certain limitations in its effectiveness. Therefore, early identification and monitoring of the development and progression of caries in children, so that targeted, comprehensive prevention and control measures can be implemented, is urgently needed.

[0003] Preschoolers often face fear or lack of cooperation, making dental caries inspections and necessary group monitoring difficult. However, saliva offers inherent advantages such as easy, painless, and non-invasive sampling, making it more user-friendly and comfortable for preschoolers, facilitating their trust and cooperation. Saliva also provides a comprehensive and integrated picture of oral health. Testing for important biomarkers, such as proteins / peptides and microorganisms, is a crucial avenue for early detection and diagnosis of dental caries, and a key tool for rapid dental caries monitoring, risk assessment, and family self-care. The composition of oral proteins and metabolites, as represented by saliva, varies significantly with age and often precedes the onset of dental caries symptoms. This suggests that meticulous monitoring and analysis of oral proteins and metabolites in children is necessary to accurately predict dental caries risk. Furthermore, many parents and the public lack awareness of dental caries prevention in children and do not understand the importance of early detection and early warning. This results in many children seeking medical attention only when dental caries symptoms become apparent, missing the optimal time for prevention and intervention. Currently, due to the uneven distribution of medical resources and the lack of professional dental specialists and necessary medical equipment in some areas, especially in remote mountainous areas, regular dental examinations and caries risk assessments are difficult to implement. Current caries risk assessment models cannot accurately predict whether an individual will develop caries within a certain period of time, and there is also a lack of quantitative analysis and assessment methods. As a result, early detection and early warning of caries risk in young children are difficult to achieve.

[0004] The existing dental caries diagnosis model based on salivary biochemical testing collects saliva samples, compares the differences in pH, total protein, and ion concentration, analyzes the correlation between salivary biochemical indicators and dental caries status, and establishes a dental caries diagnosis model. This model can distinguish between young children with dental caries and healthy children with a certain degree of accuracy. However, this model can only diagnose after dental caries occur, but cannot detect and prevent dental caries before they occur. It still cannot meet the needs of comprehensive prevention and control that the field of oral preventive medicine wants to achieve by moving the threshold forward. The currently more commonly used dental caries (Cariostat) detection method is to assess the risk of dental caries by the color corresponding to the pH value of the culture medium, and cannot perform quantitative analysis. In addition, this method is subjectively evaluated by the examiner, which has certain requirements on the technical level of the operator, and the sensitivity and specificity of the detection still have room for improvement.

[0005] Multi-omics analysis involves normalizing and comparing data sources from different omics groups, establishing relationships between data, and integrating multi-omics data to comprehensively and deeply interpret biological processes at the genetic, transcriptional, protein, and metabolic levels, thereby providing a more comprehensive understanding of biological systems. Identifying multi-omics markers associated with dental caries in young children will further understand the pathogenesis of dental caries and is also crucial for early detection and intervention of the disease. Summary of the Invention

[0006] In light of this, the present invention aims to provide a marker combination, combined reagent, and kit for early childhood caries risk assessment, as well as their use in constructing a risk assessment model for early childhood caries. This invention focuses on painless, non-invasive, and easily accessible saliva samples. Through multi-omics analysis techniques such as proteomics and metabolomics, the invention detects and screens salivary biomarkers and functional pathways associated with the dynamic changes in caries in young children. This model, with excellent stability, sensitivity, and specificity, is constructed. This model enables rapid detection, risk assessment, and family self-care of caries in young children, laying a solid foundation for effective intervention measures to achieve comprehensive caries prevention and control.

[0007] In order to achieve the above-mentioned object of the invention, the present invention provides the following technical solutions:

[0008] The present invention provides a marker combination for early risk assessment of dental caries in young children, wherein the marker combination comprises 2-oxoadipate, enoyl-CoA hydratase and kynureninase.

[0009] The present invention provides a combined reagent for early-stage dental caries risk assessment in young children. The combined reagent comprises reagents for detecting abundance reference indicators of 2-oxoadipate, enoyl-CoA hydratase, and kynureninase.

[0010] Preferably, the abundance reference index of 2-oxoadipate is obtained by a metabolomics technique based on high-throughput mass spectrometry, and the abundance reference indexes of enoyl-CoA hydratase and kynureninase are obtained by DIA proteomics detection.

[0011] Preferably, the abundance reference index is obtained based on the base 2 logarithmic value of the sample ion current intensity during the detection process.

[0012] The present invention provides a kit for early-stage caries risk assessment in young children, comprising the above-mentioned combined reagent.

[0013] The present invention also provides the use of the marker combination, the combined reagent or the kit in constructing an early risk assessment model for dental caries in young children.

[0014] Preferably, the method for constructing an early risk assessment model for dental caries in young children comprises the following steps:

[0015] (1) Collect samples and pre-process them;

[0016] (2) Proteomics analysis using liquid chromatography-mass spectrometry coupled with data-independent acquisition; non-targeted metabolomics analysis using liquid chromatography-mass spectrometry coupled with the Q-Exactive mass spectrometry platform;

[0017] (3) Substitute the proteomics and non-targeted metabolomics analysis results into the saliva multi-omics marker analysis model decision tree to obtain an early caries risk assessment model for young children:

[0018] When the abundance reference index of 2-oxoadipic acid was greater than 24.46, the abundance reference index of enoyl-CoA hydratase was greater than 7.91, and the abundance reference index of kynureninase was greater than 7.23, the risk of dental caries was high; the rest were not at high dental caries risk.

[0019] Preferably, the sample is saliva.

[0020] Preferably, the pretreatment method is to centrifuge the sample to obtain the supernatant.

[0021] Preferably, the centrifugal temperature is 2-6° C., the centrifugal speed is 8000-12000 g, and the centrifugal time is 5-15 min.

[0022] Compared with the prior art, the present invention has the following beneficial effects:

[0023] (1) The present invention uses proteomics and metabolomics to study saliva samples of young children and finds that 2-oxoadipate, enoyl-CoA hydratase, and kynureninase have a significant impact on the occurrence of dental caries in young children, indicating that 2-oxoadipate, enoyl-CoA hydratase, and kynureninase can be used as a marker combination to assess the early risk of dental caries in young children. The present invention constructs a decision tree for the saliva multi-omics marker analysis model by determining the abundance reference index of 2-oxoadipate, enoyl-CoA hydratase, and kynureninase. Based on the decision tree, the early risk of dental caries in young children can be assessed.

[0024] (2) The present invention validated the dental caries risk assessment model constructed by the present invention based on a longitudinal follow-up study with a large sample size and minimal influence from individual factors. The results showed that the dental caries risk assessment model constructed by the present invention has good test efficiency, providing new ideas for constructing a dental caries risk assessment model for young children.

[0025] (3) The dental caries risk assessment model constructed by the present invention has good stability, sensitivity, and specificity. Furthermore, the dental caries risk assessment model constructed by the present invention can achieve early diagnosis of dental caries in young children through rapid, more accurate, and effective detection and analysis, taking forward-moving prevention and control measures. This helps improve and optimize family self-care methods and lays the foundation for the development of precise prevention and control strategies for dental caries in young children. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 It is the overall flow chart of the technical solution of the present invention;

[0027] Figure 2 Decision tree for the analysis model of salivary multi-omics markers;

[0028] Figure 3 Flowchart for proteomics assay;

[0029] Figure 4 Flowchart for non-targeted metabolomics assay;

[0030] Figures 5 to 7 A multi-omics analysis combining proteomics and metabolomics was performed on saliva samples, including Figure 5 The tryptophan-kynurenine metabolic pathway was a significantly enriched pathway involved in the changes of dental caries status; Figure 6 The metabolites 2-oxoadipate, protein enoyl-CoA hydratase and kynureninase in the tryptophan-kynurenine metabolic pathway and their changing trends in different caries states; Figure 7 The receiver operating characteristic curve and area under the curve of the salivary multi-omics marker analysis model applied to early caries risk assessment in young children. DETAILED DESCRIPTION

[0031] The present invention provides a marker combination for early risk assessment of dental caries in young children, wherein the marker combination comprises 2-oxoadipate, enoyl-CoA hydratase and kynureninase.

[0032] The present invention provides a combined reagent for early-stage dental caries risk assessment in young children. The combined reagent comprises reagents for detecting abundance reference indicators of 2-oxoadipate, enoyl-CoA hydratase, and kynureninase.

[0033] In the present invention, the abundance reference index of 2-oxoadipic acid is obtained by a metabolomics technology based on high-throughput mass spectrometry, and the abundance reference indexes of enoyl-CoA hydratase and kynureninase are obtained by DIA proteomics detection; the abundance reference index is obtained based on the logarithmic value of the sample ion current intensity with a base of 2 during the detection process, and can reflect the relative expression levels of proteins and metabolites.

[0034] The present invention provides a kit for early risk assessment of dental caries in young children, the kit comprising the combination reagent, which is a reagent for detecting the abundance reference index of 2-oxoadipate, enoyl-CoA hydratase and kynureninase.

[0035] The present invention also provides the use of the marker combination, the combined reagent or the kit in constructing an early risk assessment model for dental caries in young children.

[0036] In the present invention, the method for constructing an early risk assessment model for dental caries in young children comprises the following steps:

[0037] (1) Collect samples and pre-process them;

[0038] (2) Proteomics analysis using liquid chromatography-mass spectrometry coupled with data-independent acquisition; non-targeted metabolomics analysis using liquid chromatography-mass spectrometry coupled with the Q-Exactive mass spectrometry platform;

[0039] (3) Substitute the proteomics and non-targeted metabolomics analysis results into the saliva multi-omics marker analysis model decision tree to obtain an early caries risk assessment model for young children:

[0040] When the abundance reference index of 2-oxoadipic acid was greater than 24.46, the abundance reference index of enoyl-CoA hydratase was greater than 7.91, and the abundance reference index of kynureninase was greater than 7.23, the risk of dental caries was high; the rest were not at high dental caries risk.

[0041] In the present invention, samples are collected and pretreated. The method for collecting samples is to collect saliva samples from preschool children more than 1 hour after breakfast, rinse the mouth with clean water before collecting saliva, and discard any blood stains; collect 1 to 5 mL of whole saliva, preferably 2 to 4 mL, and more preferably 3 mL; mark the sample number and place it in a centrifuge tube rack, store it in a foam plastic box with an ice pack or crushed ice, and transport it to the laboratory as soon as possible for centrifugation pretreatment; the method for pretreating the sample is to set the centrifuge to precool to 2 to 6 ° C, preferably 3 to 5 ° C, and more preferably 4 ° C; the centrifugal speed is 8000 to 12000 g, preferably 9000 to 11000 g, and more preferably 10000 g; the centrifugal time is 5 to 15 min, preferably 7 to 13 min, and more preferably 10 min, take the saliva supernatant, and put the saliva supernatant into a 1.5 mL ep tube and a 200 μL ep tube respectively.

[0042] In the present invention, liquid chromatography-mass spectrometry technology based on data-independent acquisition is used for proteomic detection; the Q-Exactive mass spectrometry detection platform is used to perform non-targeted metabolomics detection using liquid chromatography-mass spectrometry technology; the present invention uses liquid chromatography-mass spectrometry technology based on data-independent acquisition to perform proteomic detection on pretreated saliva samples, and can obtain reference indicators of the abundance of enoyl-CoA hydratase and kynurenine in the pretreated samples; the Q-Exactive mass spectrometry detection platform is used to perform non-targeted metabolomics detection using liquid chromatography-mass spectrometry technology, and can obtain a reference indicator of the abundance of 2-oxoadipic acid in the pretreated samples.

[0043] In the present invention, the detection and analysis results of proteomics and non-targeted metabolomics are substituted into the saliva multi-omics marker analysis model decision tree to obtain an early risk assessment model for dental caries in young children; the saliva multi-omics marker analysis model decision tree is as follows: Figure 2 As shown in the figure, when the abundance reference index of 2-oxoadipic acid is greater than 24.46, the abundance reference index of enoyl-CoA hydratase is greater than 7.91, and the abundance reference index of kynureninase is greater than 7.23, the risk of dental caries is high; the rest are not at high dental caries risk.

[0044] The technical solutions provided by the present invention are described in detail below with reference to the embodiments, but they should not be construed as limiting the scope of protection of the present invention.

[0045] Example

[0046] Saliva samples were collected from preschool children between 9:00 AM and 11:00 AM, ensuring that the collection time was at least 1 hour after breakfast. Before collecting saliva, rinse the mouth with clean water and discard any blood. Approximately 3 mL of whole saliva was collected, labeled, and placed in a centrifuge tube rack. The tubes were stored in a Styrofoam box with ice packs or crushed ice and transported to the laboratory as soon as possible for centrifugation pretreatment. Pretreatment was performed by precooling the centrifuge to 4°C, setting the speed to 10,000 g, and centrifuging for 10 minutes. The saliva supernatant was transferred to a 1.5 mL eppendorf tube and a 200 μL eppendorf tube, respectively, to obtain the pretreated saliva samples.

[0047] The pre-treated saliva samples were subjected to proteomic analysis using liquid chromatography-mass tandem spectrometry (LC-MS / MS) technology based on data independent acquisition (DIA) (see the operation flow for details). Figure 3) to obtain the abundance reference indicators of enoyl-CoA hydratase and kynureninase in the pretreated samples; non-targeted metabolomics detection was performed using liquid chromatography-mass spectrometry technology using the Q-Exactive mass spectrometry detection platform (operation process see Figure 4 ), and obtain the abundance reference index of 2-oxoadipic acid in the pretreated sample, the specific method is as follows:

[0048] 1. Data-independent acquisition of liquid chromatography-mass spectrometry for proteomics detection

[0049] 1.1 Sample preparation

[0050] Sample pretreatment includes protein extraction, denaturation, reductive alkylation, enzymatic digestion, and peptide desalting. The iST Sample Pretreatment Kit (PreOmics, Germany) was used to pretreat tissue samples. After grinding with liquid nitrogen, an appropriate amount of sample was added, 50 μL of lysis buffer was added, and the sample was heated at 95°C, 1000 rpm, for 10 minutes. The sample was cooled to room temperature, and trypsin digestion buffer was added. The sample was incubated at 37°C, 500 rpm, and agitation was maintained for 2 hours. The enzymatic digestion reaction was terminated by adding stop buffer. Peptide desalting was performed using the iST cartridge included in the kit, and elution was performed using 2 × 100 μL of elution buffer. The eluted peptides were vacuum-evacuated and stored at -80°C.

[0051] 1.2 Establishing a spectral database

[0052] 1) High pH reverse phase separation

[0053] All sample peptide mixtures were redissolved in buffer A (buffer A: 20 mM ammonium formate in water, adjusted to pH 10.0 with ammonia) and separated at high pH using an Ultimate 3000 system (ThermoFisher scientific, MA, USA) connected to a reversed-phase column (XBridge C18 column, 4.6 mm × 250 mm, 5 μm, Waters Corporation, MA, USA). Separation used a linear gradient from 5% B to 45% B (B: 80% ACN in 20 mM ammonium formate, adjusted to pH 10.0 with ammonia) over 40 min. The column was equilibrated at initial conditions for 15 min, with a flow rate of 1 mL / min and a column temperature of 30°C. Six fractions were collected and dried in a vacuum concentrator until use.

[0054] 2) Low pH nano-HPLC-MS / MS analysis (DDA qualitative library construction)

[0055] Desalted, lyophilized peptides were reconstituted in solvent A (0.1% formic acid in water) and analyzed by LC-MS / MS equipped with an online nanospray source. The system consisted of an Orbitrap Lumos mass spectrometer (Thermo Fisher Scientific, MA, USA) coupled to an EASY-nLC 1200 system. A total of 3 μL of sample was loaded onto an Acclaim PepMap C18 analytical column, 75 μm × 25 cm, and the sample was separated using a 120-min gradient from 5% B to 35% B (B: 0.1% formic acid in ACN). The column flow rate was controlled at 200 nL / min, and the electrospray voltage was 2 kV.

[0056] The Orbitrap Lumos mass spectrometer was operated in data-dependent acquisition mode, automatically switching between MS and MS / MS acquisition. The mass spectrometry parameters were set as follows: (1) MS: scan range (m / z): 350–1500; resolution: 120,000; AGC target: 4e5; maximum injection time: 50 ms; dynamic exclusion time: 30 s; (2) HCD-MS / MS: resolution: 15,000; AGC target = 5e4; maximum injection time: 35 ms; collision energy: 32.

[0057] 3) Search the database

[0058] Raw data were merged, analyzed, and searched using Spectronaut X (Biognosys AG). Uniprot or a provided database was used as the database. A contamination sequence library was also searched to determine if samples were contaminated. Trypsin digestion was performed. Search parameters included fixed modification: carbamidomethyl (C), variable modification: methionine oxidation. The false positive rate (FDR) was set to 1% for both the precursor ion and peptide levels.

[0059] 1.3DIA Data Acquisition

[0060] Each sample was suspended in 30 μL of solvent A (A: 0.1% formic acid in water). 9 μL was removed and added to 1 μL of 10× iRT peptides. After mixing, the mixture was separated by nano-LC and analyzed by online electrospray tandem mass spectrometry. The entire experimental system was an Orbitrap Lumos mass spectrometer (Thermo Fisher Scientific, MA, USA) connected to an EASY-nLC 1200 system. A total of 3 μL of sample was loaded onto an Acclaim PepMap C18 analytical column, 75 μm × 25 cm. Separation was performed using a 120-min gradient from 5% B to 35% B (B: 0.1% formic acid in ACN). The column flow rate was controlled at 200 nL / min, and the electrospray voltage was 2 kV.

[0061] The mass spectrometry parameters were set as follows:

[0062] (1) MS: Scan range (m / z): 350-1500; Resolution: 120,000; AGC target: 4e6; Maximum injection time: 50 ms; (2) HCD-MS / MS: Resolution: 30,000; AGC target: 1e6; Collision energy: 32; Energy increment: 5%; (3) Variable window acquisition, 60 windows were set, and overlapping serial ports were set, with each window overlapping 1 m / z.

[0063] 2. Using the Q-Exactive mass spectrometry platform, liquid chromatography-mass spectrometry was used for non-targeted metabolomics detection.

[0064] 2.1 Metabolite extraction

[0065] 1) Pipette 100 μL of sample into a 2 mL centrifuge tube;

[0066] 2) Add 400 μL of pre-chilled methanol:acetonitrile (1:1, v / v) and vortex for 30 seconds;

[0067] 3) Place in a -20°C freezer and freeze for 30 minutes;

[0068] 4) Centrifuge at 12000 rpm and 4°C for 10 min, take 400 μL of the supernatant and concentrate to dryness under vacuum;

[0069] 5) Add 150 μL of 50% methanol (containing 5 ppm 2-chlorophenylalanine) to reconstitute the solution and vortex for 30 seconds.

[0070] 6) Centrifuge at 12,000 rpm and 4°C for 10 min, filter the supernatant through a 0.22 μm filter membrane, and add the filtrate to the test bottle;

[0071] 7) Take 10-20 μL of each sample filtrate and mix them into a QC sample for evaluating instrument stability and data reliability.

[0072] 2.2 Chromatographic method:

[0073] Using ACQUITY UPLCH SS T3 Columns The flow rate was 0.4 mL / min, the column temperature was 40 °C, the autosampler temperature was 8 °C, and the injection volume was 2 μL.

[0074] Positive and negative mode mobile phase: mobile phase A is 0.1% formic acid water, mobile phase B is acetonitrile (containing 0.1% formic acid), elution gradient is shown in Table 1:

[0075] Table 1 Elution gradient in positive and negative modes

[0076]

[0077]

[0078] 2.3 Mass spectrometry method:

[0079] DDA mass spectrometric data were acquired in both positive and negative ion modes using a Thermo Orbitrap Exploris 120 mass spectrometer controlled by Xcalibur software (version 4.7, Thermo). A HESI source was used, with a spray voltage of 3.5 kV / -3.0 kV, a sheath gas of 40 arb, an auxiliary gas of 15 arb, a capillary temperature of 325°C, an auxiliary gas temperature of 300°C, a primary resolution of 60,000, a scan range of 100–1000 m / z, an AGC target of standard, a Max IT of 100 ms, and secondary fragmentation with the top four ions selected for screening. A dynamic exclusion time of 8 s was used, a secondary resolution of 15,000, an HCD collision energy of 30%, an AGC target of standard, and a Max IT of Auto.

[0080] All samples to be tested and QC samples were loaded into the instrument according to the above-mentioned chromatographic and mass spectrometric methods. Before the formal injection, 2 to 4 QC samples were injected to balance the system. During the injection process, one QC sample was injected for every 5 to 10 samples for subsequent data evaluation and quality control.

[0081] 3. Obtaining Proteomic Analysis Results

[0082] First, quality control analysis was performed on the original data (Raw Data) corresponding to the samples. On this basis, differential proteins were screened, and principal component analysis, correlation analysis, expression pattern cluster analysis, etc. were performed on the difference comparison group data.

[0083] Then, functional annotation analysis of the differentially expressed proteins was performed, including GO analysis, KEGG Pathway analysis, protein interaction analysis, etc. Through this series of analyses, key proteins and their functions or pathways were selected.

[0084] 4. Obtaining metabolomics analysis results

[0085] 4.1 Metabolite library search

[0086] Import the downloaded .raw format data into the commercial software Compound Discoverer TM3.3 (version 3.3.2.31, Thermo, Waltham, USA), based on the software's new peak detection and peak quality scoring algorithm, performs peak extraction, alignment, correction and other operations. The unique peak quality rating calculation and filter (Peak rating calculation and filter) greatly reduces the interference of background peaks and low-quality peaks. Peaks that are not detected in more than 50% of QC samples are filtered, and the undetected peaks are filled with missing values ​​based on the software's Fill Gaps algorithm, and the Sum total peak area is normalized. The identification of metabolites is based on self-built libraries, mzCloud online libraries (https: / / www.mzcloud.org / ), LIPID MAPS (https: / / www.lipidmaps.org / ), HMDB (https: / / hmdb.ca / ), MoNA (https: / / mona.fiehnlab.ucdavis.edu / ) and NIST_2020_MSMS spectral library. The MS1 mass tolerance MassTolerance is set to 15ppm and the MS2 Match FactorThreshold is set to 50.

[0087] 4.2 Bioinformatics Analysis

[0088] (1) Expression abundance analysis

[0089] The metabolite abundance values ​​were plotted as expression abundance density maps and violin plots using ggplot2 (V3.4.1). The metabolite abundance values ​​were clustered using the Pheatmap package (V1.0.12) in R. At the same time, two-way clustering was performed on samples and metabolites, and a heat map was drawn to display the abundance of sample metabolites.

[0090] (2) Comparative analysis of differences between the two groups

[0091] The R software package Ropls was used to perform principal component analysis (PCA), partial least squares discriminant analysis (PLS-DA), and orthogonal partial least squares discriminant analysis (OPLS-DA) dimensionality reduction analyses on the sample data. Score, loading, and S-plots were plotted to demonstrate differences in metabolite composition between samples. The model was tested for overfit using a permutation test. R²X and R²Y represent the explanatory power of the model for the X and Y matrices, respectively, and Q² indicates the predictive power of the model. Values ​​closer to 1 indicate better model fit and more accurate classification of training set samples into their original categories. P-values ​​were calculated using statistical tests, variable projection importance (VIP) using the OPLS-DA dimensionality reduction method, and fold change (FC) was calculated between groups to measure the influence and explanatory power of each metabolite component on sample classification and discriminant analysis, assisting in the screening of marker metabolites. Metabolites were considered statistically significant when the p-value was <0.05 and the VIP value was >1.

[0092] The Pheatmap package (V1.0.12) in R was used to perform cluster analysis on the abundance values ​​of differential metabolites, and heat maps and trend analysis diagrams were drawn; VennDiagram (V1.7.3) and UpSetR (V1.4.0) were used to draw Venn diagrams and Upset diagrams for the differential substances compared between two different groups; corrplot (V4.0.3) was used to perform correlation analysis on the differential metabolites; ggplot2 (V3.4.1) was used to draw box plots and violin plots of differential metabolites to show the abundance of each differential substance between different groups; the differential metabolite results were further analyzed by machine learning (mlr3verse, V0.2.7) and ROC curve drawing (pROC, V1.18.2) to obtain key product information in the differential set.

[0093] Functional analysis of differential metabolites was performed, mainly by performing KEGG enrichment analysis on differential substances through clusterProfiler (V4.6.0) to obtain significantly enriched metabolic pathway information, and by calculating the differential abundance score, the overall change differential abundance score of all differential metabolites in a certain pathway was obtained, thereby capturing the average and overall change trend of all metabolites in a certain pathway and better screening key pathways.

[0094] (3) Multi-group comparison and difference analysis

[0095] Multi-group comparison statistical analysis was performed using PMCMRplus (V1.9.6) to obtain significant p-values. Principal component analysis (PCA) and partial least squares discriminant analysis (PLS-DA) were performed on the sample data using the R software package Ropls. Score, loading, and S-plot plots were generated to demonstrate differences in metabolite composition between samples. The model was tested for overfitting using a permutation test. R2X and R2Y represent the explanatory power of the model for the X and Y matrices, respectively, and Q2 indicates the predictive power of the model. Values ​​closer to 1 indicate a better fit of the model and a more accurate classification of the training set samples into their original categories. P-values ​​were calculated based on statistical tests; when the p-value was < 0.05, differences in metabolite molecules between groups were considered statistically significant.

[0096] Cluster analysis of differential metabolite abundance values ​​was performed using the Pheatmap package (V1.0.12) in R, with heatmaps and trend analysis plots drawn. Venn diagrams and Upset plots were drawn for differential substances in different multi-group comparisons using VennDiagram (V1.7.3) and UpSetR (V1.4.0). Correlation analysis of differential metabolites was performed using corrplot (V4.0.3). The differential metabolite results were further analyzed using machine learning analysis (mlr3verse, V0.2.7) and receiver operating characteristic (ROC) curve drawing (pROC, V1.18.2) to obtain key product information in the differential set. Functional analysis of differential metabolites was performed, primarily using clusterProfiler (V4.6.0) for KEGG enrichment analysis of differential substances to obtain significantly enriched pathway information.

[0097] According to the proteomics and metabolomics analysis results obtained in the study, the saliva multi-omics marker analysis model decision tree ( Figure 2 ), combined with the abundance reference indicators of three key markers: the saliva metabolite 2-oxoadipate, the saliva protein enoyl-CoA hydratase and kynureninase. The abundance value of the saliva metabolite 2-oxoadipate was obtained by high-throughput mass spectrometry-based metabolomics technology, and the saliva protein enoyl-CoA hydratase and kynureninase were obtained by DIA proteomics technology, so as to determine whether the individual from which the sample was obtained has a high risk of developing dental caries in young children.

[0098] The present invention carried out a 6-month validation cohort to apply the multi-omics marker analysis model to the early risk assessment of dental caries in young children. The specific process is shown in Figure 1 A total of 103 preschool children were included in the cohort. They were followed up every 3 months, examined for caries, and their caries status was recorded. Saliva samples were collected and multi-omics analysis combining proteomics and metabolomics was performed. The results are shown in Figures 5 to 7 .

[0099] The study found that the expression of the saliva metabolite 2-oxoadipic acid, the saliva protein enoyl-CoA hydratase, and the kynureninase all showed a significant upward trend during the stable period of filling treatment, which was a completely different trend compared with the children in the study cohort who suffered from caries again (i.e., new caries, secondary caries, or recurrent caries) at the next follow-up time point after caries treatment. The detection results of the three components of the saliva metabolite 2-oxoadipic acid, the saliva protein enoyl-CoA hydratase, and the kynureninase were combined, and the receiver operating characteristic (ROC) curve was drawn based on the decision tree of the salivary multi-omics marker analysis model. The area under the curve (AUC) value of the three-combination model reached a level exceeding 0.90 (the actual AUC value was 0.900826, as shown in Figure 2). Figure 7 The results showed that the salivary multi-omics biomarker analysis model constructed based on three salivary biomarkers, namely the salivary metabolite 2-oxoadipic acid, salivary protein enoyl-CoA hydratase, and salivary kynurenine enzyme, has good test efficiency in the application of early caries risk assessment in young children, helps to identify the occurrence and prognosis of caries, and has important practical significance and application value.

[0100] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as within the scope of protection of the present invention.

Claims

1. A marker combination for early risk assessment of dental caries in young children, characterized in that: The marker combination consists of 2-oxoadipate, enoyl-CoA hydratase and kynureninase.

2. Use of the marker combination according to claim 1 in constructing an early risk assessment model for dental caries in young children.

3. The use according to claim 2, characterized in that The method for constructing an early risk assessment model for dental caries in young children comprises the following steps: (1) Collect samples and pre-process them; (2) Proteomics detection using liquid chromatography-mass spectrometry coupled with data-independent acquisition; non-targeted metabolomics detection using liquid chromatography-mass spectrometry coupled with the Q-Exactive mass spectrometry platform; (3) Substitute the results of proteomics and non-targeted metabolomics into the decision tree of the saliva multi-omics marker analysis model to obtain an early risk assessment model for dental caries in young children: When the abundance reference index of 2-oxoadipic acid was greater than 24.46, the abundance reference index of enoyl-CoA hydratase was greater than 7.91, and the abundance reference index of kynureninase was greater than 7.23, the risk of dental caries was high; the rest were not at high dental caries risk.

4. The use according to claim 3, characterized in that The sample is saliva.

5. The use according to claim 3, characterized in that The pretreatment method is to centrifuge the sample to obtain the supernatant.

6. The use according to claim 5, characterized in that The centrifugal temperature is 2-6° C., the centrifugal speed is 8000-12000 g, and the centrifugal time is 5-15 min.

Images