Oral cavity microorganism for identifying taste sensitivity as well as identification method and application of oral cavity microorganism

Abstract

The invention belongs to the technical field of microorganisms, and particularly relates to an oral microorganism for identifying taste sensitivity as well as an identification method and application of the oral microorganism. On the basis of comparative study of young and old population, a screened oral microorganism combination covers wide characteristics of taste sensitivity difference, and the abundance of one or more of five oral microorganisms of a Haemophilus genus, a Fusobacterium genus, an Aggregate genus, a Lacnoanaerobacter genus and an Oribacterium genus is in significant positive correlation with the taste sensitivity, so that the oral microorganism combination can be used for accurately identifying the oral microorganisms of one or more of the five oral microorganisms of the Haemophilus genus, the Fusobacterium genus, the Aggregate genus, the Lacnoanaerobacter genus and the Oribacterium genus.

Description

Technical Field

[0001] The present invention belongs to the technical field of microorganisms, and in particular relates to an oral microorganism for identifying taste sensitivity and an identification method and application thereof. Background Art

[0002] Food perception is a person's response and feedback to food stimuli, which in turn affects food selection, acceptance, and intake. Differences in food perception are one of the driving factors affecting food consumption, and are also a regulatory factor that affects people's dietary needs and personal preferences. As the starting point of the digestive system, the oral cavity provides an important ecological site for the interaction and accumulation of food particles and microorganisms, directly affecting an individual's ability to perceive the taste of food. Among them, oral microorganisms are the second largest microbial community in the human body and have powerful metabolic capabilities. Oral microorganisms can affect the host's sensitivity to taste perception, regulate the host's sensory decisions, and influence their food choices.

[0003] At present, there are generally few literatures on oral microorganisms as microbial markers in the patents published at home and abroad, and almost no literature involves the early diagnosis of taste sensitivity.

[0004] (1) The application of oral microorganisms as microbial markers is mainly concentrated in the medical field. The published patents focus on diabetes, obesity, esophageal cancer, lung cancer, COVID-19, neuropsychiatric symptoms, etc.: constructing a population cohort (experimental group and healthy control group), collecting saliva samples, extracting DNA, sequencing and data analysis, and using conventional statistical analysis and other theoretical demonstration methods as identification methods to screen key oral microorganisms. Among these related patents, some patents use machine learning algorithms such as random forests and logistic regression instead of conventional statistical analysis to screen key oral microorganisms from a single taxonomic level (genus or species).

[0005] (2) Regarding methods for determining taste sensitivity, a patent has been published regarding methods and compositions for diagnosing and treating loss and / or distortion of taste or smell that is similar: the degree of loss of taste or smell in an individual is diagnosed by measuring the level of Sonic Hedgehog in nasal mucus, saliva, or a combination thereof.

[0006] Conventional statistical methods can identify oral microbes with significant differences between groups, but the oral cavity is a complex ecosystem with a large amount of microbial diversity data, making it difficult to accurately identify key microbes using conventional statistical methods alone. Machine learning algorithms can handle complex data well, but are limited to the level of a single species. This paper combines differential analysis methods with machine learning algorithms to select the model with better classification performance from four candidate classification models: logistic regression, linear support vector machine, Bayesian ridge regression, and Gaussian process classifier, to accurately identify the key characteristic oral microbes used to determine the taste sensitivity of different people.

[0007] Using sonic hedgehog protein levels to assess sensory loss in individuals—using the protein as a marker—is effective, but antibody-based ELISA methods are prone to cross-reactions, dimers, and other false positives, leaving room for improvement in the accuracy of assessing taste deterioration in large populations. The oral microbiome selected in this study, using DNA-based microbial markers combined with machine learning algorithms to identify key oral microbes, has demonstrated high accuracy in assessing taste sensitivity. Summary of the Invention

[0008] Currently, there is a lack of oral microbial combinations and identification methods that are specifically associated with taste sensitivity. In response to this gap, this patent application provides an oral microbial combination for identifying taste sensitivity and a screening method thereof. Since there is significant heterogeneity in taste sensitivity among different age groups (taste sensitivity generally decreases with age), selecting individuals from two age groups, young people and the elderly, as research subjects can cover a wide range of characteristics of taste differences, thereby enhancing the accuracy and universality of screening oral microbial combinations that are significantly associated with taste sensitivity. By comparing the differences in taste sensitivity between individuals of different age groups, accurately identifying characteristic oral microbial combinations based on differential analysis and combining machine learning algorithms, and judging individual taste sensitivity differences through non-invasive and non-invasive means, it aims to achieve efficient, low-cost, and accurate identification of individual taste sensitivity, and provide a reliable basis for personalized nutritional intervention.

[0009] In view of this, the present invention provides an oral microbiome combination for identifying taste sensitivity and its application, comparing the taste sensitivity differences among individuals of different age groups, accurately identifying characteristic oral microbiome combinations using differential analysis and machine learning algorithms, and determining individual taste sensitivity differences through non-invasive and non-invasive means. To achieve the above-mentioned invention objectives, the present invention adopts the following technical solutions:

[0010] The first aspect of the present invention is to provide an oral microbial combination for taste sensitivity, characterized in that the microorganisms include one or more of the genera Haemophilus, Fusobacterium, Aggregatibacter, Lachnoanaerobaculum and Oribacterium, wherein the abundance of the one or more microorganisms is significantly positively correlated with taste sensitivity.

[0011] Furthermore, the taste includes one or more of sour, sweet, bitter, salty and umami.

[0012] A second aspect of the present invention is to provide a method for identifying taste sensitivity of a subject, characterized in that the method comprises:

[0013] 1) Collect oral biological samples;

[0014] 2) Microbial sample testing: DNA is extracted from saliva samples and sequenced;

[0015] 3) detecting the abundance of the microbial combination as described in the first aspect;

[0016] 4) Verify the correlation between key characteristic oral microbiota and individual taste sensitivity.

[0017] Furthermore, the abundance of the detected microorganisms is: 0.05% to 20.42% for Haemophilus, 0% to 11.4% for Fusobacterium, 0% to 3.76% for Aggregatibacter, 0% to 1.22% for Lachnoanaerobaculum, and 0% to 3.88% for Oribacterium.

[0018] Furthermore, the taste includes one or more of sour, sweet, bitter, salty and umami.

[0019] Furthermore, the sequencing in step 2) is second-generation or third-generation high-throughput sequencing.

[0020] Furthermore, the method is applied in non-disease diagnosis.

[0021] The third aspect of the present invention is to provide an oral microbial combination for identifying taste sensitivity and its application, specifically comprising:

[0022] (1) Data collection: Based on age differences, we recruited elderly and young subjects, tested their taste sensitivity, and collected oral biological samples.

[0023] (2) Data conversion: extract DNA from oral biological samples and use bioinformatics methods based on sequencing analysis to convert DNA information into oral microbial community information;

[0024] (3) Screening of key oral microorganisms: preliminary screening of key oral microorganisms based on differential analysis.

[0025] (4) Identification of key characteristic oral microorganisms: Based on the characterization of individual taste sensitivity, a model with better classification effect is selected from multiple candidate classification models of machine learning algorithms and hyperparameter optimization is performed. The optimized model is used to identify characteristic oral microorganisms, and the key oral microorganisms obtained in (3) are further combined to obtain key characteristic oral microorganisms.

[0026] (5) Verification of key characteristic oral microorganisms: Analyze the correlation between key characteristic oral microorganisms and individual taste sensitivity for verification analysis.

[0027] Furthermore, the oral biological sample includes one or more of a saliva sample, a tongue dorsum sample, and a supragingival dental plaque sample.

[0028] Furthermore, the taste includes one or more of sour, sweet, bitter, salty and umami.

[0029] Furthermore, the taste sensitivity includes one or more of a perception threshold, a recognition threshold, and a difference threshold.

[0030] Furthermore, the bioinformatics method includes one or more of 16s rRNA sequencing and metagenomic sequencing.

[0031] Furthermore, the candidate classification models of the machine learning algorithm include one or more of logistic regression, linear support vector machine, Bayesian ridge regression and Gaussian process classifier.

[0032] The patent application provides the following beneficial effects:

[0033] 1. Based on a comparative study of young and elderly groups, the screened oral microbiome combination covers a wide range of characteristics of taste sensitivity differences and has good accuracy and universality.

[0034] 2. It adopts a non-invasive sample collection method, which is easy to operate and meets the needs of large-scale population screening.

[0035] 3. Through differential analysis combined with machine learning algorithms, five oral microorganisms were accurately identified for identifying individual taste sensitivity, and all showed a significant positive correlation with taste sensitivity. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 Specific implementation plan overall methods and steps;

[0037] Figure 2 Results of sweetness perception and recognition thresholds of people of different ages (***p<0.001, N=60);

[0038] Figure 3 Analysis of the genus-level composition of oral microbial communities in people of different ages;

[0039] Figure 4 Comparison of oral microbial alpha diversity among people of different ages;

[0040] Figure 5 UMAP map of oral microbial abundance (left) and UMAP map of sweet taste perception threshold (right);

[0041] Figure 6Results of oral microbial abundance difference analysis (left) and LEfSe analysis (right);

[0042] Figure 7 Initial classification performance of candidate classification models (left) and classification performance of LR and SVC models after hyperparameter optimization (right);

[0043] Figure 8 Key characteristic oral microorganisms identified based on LR and SVC models;

[0044] Figure 9 Validation results of the correlation between key characteristic oral microorganisms and sweetness threshold;

[0045] Figure 10 Validation results of the correlation between key characteristic oral microorganisms and salty taste threshold;

[0046] Figure 11 Validation results of the correlation between key characteristic oral microorganisms and sour taste threshold. DETAILED DESCRIPTION

[0047] The following is a further description of the concept of the present invention and the technical effects produced in conjunction with specific embodiments, so as to fully understand the purpose, features and effects of the present invention. The methods described are all conventional methods unless otherwise specified. The materials described can be obtained from public commercial channels unless otherwise specified. The illustrative embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute undue limitations of the present invention. It should be noted that, unless there is a conflict, the embodiments of the present invention and the features in the embodiments can be combined with each other.

[0048] Example 1 Microbial Screening and Identification

[0049] according to Figure 1 The process of designing and implementing experiments:

[0050] 1. Data Collection

[0051] (1) Subject recruitment

[0052] Based on age differences, 30 subjects were recruited from the young group (20-30 years old) and the elderly group (60-70 years old). Recruitment requirements: The subjects were in good health, had no smoking or heavy drinking habits, had not received specific periodontal treatment, and had not taken antibiotics in the past 3 months. In addition, the subjects were required to have no oral diseases, such as bleeding gums, oral ulcers, or toothaches, on the day of the experiment. All subjects signed the volunteer informed consent form for subsequent research. The information of all subjects is shown in Table 1.

[0053] Table 1 Basic statistics of subjects

[0054]

[0055]

[0056] (2) Collecting taste sensitivity information

[0057] The differences in individual taste sensitivity in different groups were explored based on the three-point ascending forced choice method. A 15mL taste stimulus sample (food-grade sucrose) of a certain concentration was prepared with pure water and provided to the subjects together with two 15mL reference samples (pure water). The subjects were asked to choose a cup of sample that was different from the other two cups from the three cups of samples presented and evaluate its taste attributes (choose one of seven from sour, sweet, bitter, salty, fresh, water-like taste, and indescribable taste). The test cups were exactly the same tasting cups and were marked with three random numbers. The order of presentation was implemented in an increasing concentration sequence. All samples were tasted at room temperature, and pure water was provided to clean the mouth during the period. In this embodiment, all research subjects who met the inclusion criteria were required not to eat for at least 1 hour before sampling (drinking water was allowed) and to rinse their mouths with pure drinking water before the test. The perception threshold and recognition threshold of sweetness were calculated according to the following formula:

[0058]

[0059] Among them, C is the threshold, C n is the last incorrectly identified concentration, C n+1 is the last concentration that was not correctly identified. Figure 2 As shown, the perception and recognition thresholds for the young group were 2.05±1.80g / L and 2.98±2.09g / L, respectively, while those for the elderly group were 11.76±11.50g / L and 13.53±11.15g / L, respectively. This indicates that the elderly group had significantly higher sweetness perception and recognition thresholds than the young group. Furthermore, the threshold results for both groups spanned a wide range, effectively covering the broad range of differences in taste sensitivity across different populations.

[0060] (3) Collecting oral biological samples

[0061] To avoid the influence of circadian rhythm on saliva secretion, the saliva collection time is fixed between 9:00 and 11:00 a.m. every day. The subject sits on a chair, lowers his head 45 degrees, and lightly touches the front teeth with the tip of his tongue, allowing saliva to flow out and accumulate naturally in the mouth. He gently spits saliva into the oval funnel of the sterile sampler to collect the saliva secreted by the subject in a natural resting state (non-stimulated state). In this embodiment, all subjects who meet the inclusion criteria are required to not eat for at least 1 hour before sampling (drinking water is allowed), and rinse their mouths with purified drinking water before sampling to ensure that there is no food residue in the mouth.

[0062] 2. Data conversion

[0063] DNA was extracted from saliva samples using the FastPure Soil DNA Isolation Kit. The integrity of the extracted genomic DNA was detected by 1% agarose gel electrophoresis, and the DNA concentration and purity were determined using NanoDrop2000.

[0064] The 16S rRNA gene V3-V4 variable region was amplified by PCR using the extracted DNA as a template using the upstream primer 338F (5'-ACTCCTACGGGAGGCAGCAG-3', SEQ ID NO: 1) and the downstream primer 806R (5'-GGACTACHVGGGTWTCTAAT-3', SEQ ID NO: 2). The PCR reaction system was as follows:

[0065] 5×TransStart FastPfu buffer 4 μL,

[0066] 2.5mM dNTPs 2μL,

[0067] 5μM primer 0.8μL,

[0068] TransStart FastPfu DNA polymerase 0.4 μL,

[0069] Template DNA 10ng, make up to 20μL.

[0070] The amplification procedure is as follows:

[0071] Pre-denaturation at 95°C for 3 min;

[0072] 95°C denaturation for 30 s, 55°C annealing for 30 s, and 72°C extension for 30 s, for 27 cycles;

[0073] Extension at 72°C for 10 min.

[0074] The PCR products were run on 2% agarose gel to confirm the amplification effect.

[0075] The purified PCR products were constructed using the NEXTFLEX Rapid DNA-Seq Kit and sequenced using the Illumina PE300 platform. Sequencing data were spliced and filtered to obtain high-quality sequences, which were clustered into operational taxonomic units (OUTs) for species classification at a 97% similarity level. Chloroplast and mitochondrial sequences were removed from all samples. OTU species taxonomic annotation was performed against the Silva 16S rRNA gene database to obtain information on oral microbial communities. Alpha diversity was used to analyze the richness and diversity of oral microbial communities. Results are shown in the table. Figure 3As shown: A total of 421 oral microbial genera were detected in the two groups, and the 10 most abundant genera were Streptococcus, Prevotella, Neisseria, Haemophilus, Rothia, Veillonella, Porphyromonas, Leptotrichia, Fusobacterium, and Granulicatella species. Further analysis found that there were significant differences in α-diversity indices such as the observed species index, ace index, and Chao 1 index between the two groups ( Figure 4 ), indicating that there are significant differences in the species richness of oral microorganisms between young and elderly people.

[0076] 3. Screening of key oral microorganisms

[0077] The UMAP dimensionality reduction algorithm was used to observe the distribution of oral microbial abundance information and sweet taste sensitivity information between the two groups. Figure 5 As shown in the figure, the UMAP-based classification method can clearly distinguish the oral microbial information of the two groups, and the distribution is consistent with the threshold results. At the genus level, the abundance difference analysis was used to preliminarily screen the differential microorganisms between the elderly group and the young group. From the phylum to the genus level, LEfSe analysis was used to further identify the key differential microorganisms. The results are shown in the figure. Figure 6 As shown, selected key oral microorganisms include:

[0078] The key oral microorganisms with significantly increased relative abundance in the elderly group included Actinomyces, Clostridia_UCG-014, and TM7x. The key oral microorganisms with significantly decreased relative abundance in the elderly group included Haemophilus, Acinetobacter, Fusobacterium, Aggregatibacter, Lachnoanaerobaculum, Lautropia, Oribacterium, and Neisseria.

[0079] 4. Identify key characteristics of oral microbes

[0080] Four machine learning algorithms, namely logistic regression (LR), linear support vector machine (SVC), Bayesian ridge regression (BRR) and Gaussian process classifier (GPC), were selected as candidate classification models. The classification accuracy was evaluated based on the area under the receiver operating characteristic curve (AUC ROC). The classification models with higher accuracy, namely LR and SVC ( Figure 7 A).

[0081] On this basis, the random search method is further used to optimize the hyperparameters of these two models: param_l1_ratio and param_C parameters are optimized in the LR model, and paradm_fit_intercept and param_C parameters are optimized in the SVC model.

[0082] The two optimized models both achieved high AUC ROC scores of around 0.9 ( Figure 7 B). Feature importance analysis was performed using SHAP feature selection to identify the top 10 important features of the two models (Oral Microbiome Figure 8 ). Combining the SHAP feature selection and the key oral microbial results screened in step 3, the key characteristic oral microorganisms were selected as Haemophilus, Fusobacterium, Aggregatibacter, Lachnoanaerobaculum and Oribacterium. Figure 7-8 shown.

[0083] 5. Verification of key features of oral microorganisms

[0084] The correlation between key characteristic oral microorganisms and individual taste sensitivity was analyzed for validation analysis. Among the five key characteristic oral microorganisms, Haemophilus, Fusobacterium, Aggregatibacter, and Lachnoanaerobaculum were all significantly negatively correlated with the sweet taste perception threshold (p < 0.05), that is, they were significantly positively correlated with sweet taste sensitivity. The results are as follows: Figure 9 This suggests that the key characteristic oral microorganisms identified by the method can be used to judge individual sweet taste sensitivity.

[0085] Example 2 Salty taste sensitivity verification

[0086] The only difference from Example 1 is that food-grade sodium chloride was used as the taste stimulus sample for the collection of taste sensitivity information and the salty taste threshold was calculated. The correlation between the key characteristic oral microorganisms and individual salty taste sensitivity was verified. The five key characteristic oral microorganisms, Haemophilus, Fusobacterium, Aggregatibacter, Lachnoanaerobaculum, and Oribacterium, were all significantly negatively correlated with the salty taste perception threshold (p < 0.05), that is, they were significantly positively correlated with salty taste sensitivity (the results are shown in Figure 2). Figure 10 This indicates that the key characteristic oral microorganisms identified based on the method can be used to judge individual salty taste sensitivity.

[0087] Example 3 Sour taste sensitivity verification

[0088] The only difference from Example 1 is that food-grade sodium citrate monohydrate was used as the taste stimulus sample for the collection of taste sensitivity information and the sour taste threshold was calculated. The correlation between the key characteristic oral microorganisms and individual sour taste sensitivity was verified. Among the five key characteristic oral microorganisms, Haemophilus, Fusobacterium, Aggregatibacter, and Lachnoanaerobaculum genera were all significantly negatively correlated with the sour taste perception threshold (p < 0.05), that is, they were significantly positively correlated with sour taste sensitivity (the results are shown in Figure 2). Figure 11 This indicates that the key characteristic oral microorganisms identified by the method can be used to judge individual sour taste sensitivity.

[0089] The embodiments described above are only part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, other embodiments obtained by those skilled in the art without creative work shall fall within the scope of protection of the present invention.

Claims

1. An oral microbial combination for taste sensitivity, characterized in that: The microorganisms include one or more of the genera Haemophilus, Fusobacterium, Aggregatibacter, Lachnoanaerobaculum and Oribacterium, wherein the abundance of the one or more microorganisms is significantly positively correlated with taste sensitivity.

2. The microbial combination according to claim 1, characterized in that The taste includes one or more of sour, sweet, bitter, salty and fresh.

3. A method for identifying taste sensitivity of a subject, characterized in that: The method described is: 1) Collect oral biological samples; 2) Microbial sample testing: DNA is extracted from saliva samples and sequenced; 3) detecting the abundance of the microbial combination according to claim 1 or 2; The abundance of one or more of the above microorganisms was significantly positively correlated with taste sensitivity.

4. The method according to claim 3, characterized in that Furthermore, the abundance of the detected microorganisms is: 0.05% to 20.42% for Haemophilus, 0% to 11.4% for Fusobacterium, 0% to 3.76% for Aggregatibacter, 0% to 1.22% for Lachnoanaerobaculum, and 0% to 3.88% for Oribacterium.

5. An oral microbial combination for identifying taste sensitivity and its application, specifically comprising: (1) Data collection: Based on age differences, we recruited elderly and young subjects, tested their taste sensitivity, and collected oral biological samples. (2) Data conversion: extract DNA from oral biological samples and use bioinformatics methods based on sequencing analysis to convert DNA information into oral microbial community information; (3) Screening of key oral microorganisms: preliminary screening of key oral microorganisms based on differential analysis; (4) Identifying key characteristic oral microorganisms: Based on the individual's taste sensitivity characterization, a model with better classification effect is selected from multiple machine learning algorithm candidate classification models and hyperparameter optimization is performed. The optimized model is used to identify characteristic oral microorganisms, and the key characteristic oral microorganisms obtained in (3) are further combined to obtain the key characteristic oral microorganisms; (5) Verification of key characteristic oral microorganisms: Analyze the correlation between key characteristic oral microorganisms and individual taste sensitivity for verification analysis.

6. The method according to claim 5, characterized in that In step (1), the taste sensitivity includes one or more of a perception threshold, a recognition threshold, and a difference threshold.

7. The method according to claim 5, characterized in that The bioinformatics method includes one or more of 16s rRNA sequencing and metagenomic sequencing.

8. The method according to claim 5, characterized in that The differential analysis in step (3) is to use abundance differential analysis to identify microbial groups with significant abundance differences between the two groups, and use LEfSe analysis to further identify key differential microorganisms.

9. The method according to claim 5, characterized in that The candidate classification models of the machine learning algorithm in step (4) include one or more of logistic regression, linear support vector machine, Bayesian ridge regression and Gaussian process classifier.

10. The method according to claim 3 or claim 7, characterized in that The sequencing described in step 2) is second-generation or third-generation high-throughput sequencing.

Images