Application of biomarker in preparation of product for distinguishing patients with dampness syndrome in healthy people

Abstract

The invention discloses application of a biomarker in preparation of a product for distinguishing patients with dampness syndromes in healthy people. The invention relates to an application of a lipid lysophosgene choline-16: 0, a lipid lysophosgene choline-22: 0, a lipid phosgene choline-22: 0, a lipid phosgene choline-26: 0, a lipid phosgene choline-34: 0 and / or a lipid Hexosylceramide-42: 2 as a moisture syndrome early warning biomarker in preparation of a product for distinguishing moisture syndrome patients in healthy people. The invention further relates to an application of the lipid lysophosgene choline-16: 0, a lipid lysophosgene choline-22: 0, a lipid phosgene choline-26: 0, a lipid phosgene choline-34: 0 and / or a lipid Hexosylceramide-42: 2 as the moisture syndrome early warning biomarker. According to the present invention, the five core lipids such as LPC 16: 0, LPC 22: 0, PC 26: 0, SM 34: 0 and HexCer 42: 2 show good curve lower peak area (AUC), accuracy, precision, sensitivity and F1 recovery rate in the discovery queue; it is indicated that five lipids including LPC 16: 0, LPC 22: 0, PC 26: 0, SM 34: 0 and HexCer 42: 2 have the potential to become the wet syndrome early warning biomarkers.

Description

Technical Field

[0001] The present invention relates to the field of biomarker technology, which is used to establish the field of biomarker determination for the diagnosis of TCM dampness syndrome patients in healthy populations according to Western medical indicators, and specifically to the application of biomarkers in the preparation of products for distinguishing dampness syndrome patients in healthy populations. Background Art

[0002] Syndrome is a unique concept in traditional Chinese medicine and occupies a core position in the diagnosis and treatment of traditional Chinese medicine. Dampness syndrome is one of the common pathological syndromes in clinical practice. In traditional Chinese medicine theory, dampness syndrome refers to a pathological state caused by the invasion of external dampness or abnormal water transportation in the body. It is often manifested as adverse reactions such as obesity, thick and greasy tongue coating, and sticky stools. Traditional Chinese medicine believes that dampness is the pathogenic factor of many diseases and is also the primary cause of various diseases.

[0003] Standardized and scientific clinical diagnosis of dampness syndrome and early warning of dampness syndrome in traditional Chinese medicine play a vital role in controlling diseases and reducing the incidence of diseases. The four diagnostic methods of traditional Chinese medicine (looking, smelling, asking, and palpating) laid the foundation for clinical diagnosis of traditional Chinese medicine. On this basis, domestic researchers have established a dampness syndrome assessment scale in traditional Chinese medicine with "dampness" as the main symptom characteristic based on the theory of traditional Chinese medicine, through the operational definition of the concept of dampness syndrome and other methods. This table covers 30 dampness syndrome-related items, each of which contains five levels: none, mild, moderate, severe, and extremely severe. The scale quantitatively evaluates the severity of dampness syndrome, with "none" corresponding to 0 points, "mild" corresponding to 1 point, "moderate" corresponding to 2 points, "severe" corresponding to 3 points, and "extremely severe" corresponding to 4 points. The total score is 120 points. The higher the score, the more severe the dampness syndrome.

[0004] The theoretical system of traditional Chinese medicine emphasizes integrity and systematicity. Looking at the composition of the items in the dampness syndrome assessment scale, it covers four major categories: 1. Oral related items: such as greasy tongue coating, thick tongue coating, heavy breath, etc.; 2. Gastrointestinal related items: such as sticky stool, unformed stool, etc.; 3. Physical characteristics related items: such as obesity, greasy face or hair, etc.; 4. Mental state related items: such as fatigue, sleepiness, etc. These items complement each other, have good representativeness and high sensitivity, ensure the scientific rationality of the clinical diagnosis method of dampness syndrome, and promote the objectivity and standardization of the clinical diagnosis of dampness syndrome. However, its regionality and the representativeness and universality of the samples need to be further studied. At present, the clinical evaluation of dampness syndrome in traditional Chinese medicine is in its infancy, and a standard and standardized evaluation model has not yet been formed. The items covered by the dampness syndrome assessment scale still lack strong experimental support and data support, and the biological connotation behind dampness syndrome is still unclear. This field urgently needs to be explored and studied.

[0005] Therefore, using TCM clinical samples as a research vehicle to identify early warning biomarkers of dampness syndrome is of great significance for improving the diagnostic accuracy and early intervention effect of dampness syndrome. Summary of the Invention

[0006] The present invention provides five biomarkers for use as biomarkers in the preparation of products for discriminating patients with damp syndrome among healthy populations.

[0007] Specifically, the present invention provides that five core lipids, Lysophosphatidyl choline-16:0 (LPC 16:0), Lysophosphatidyl choline-22:0 (LPC 22:0), Phosphatidylcholine-26:0 (PC 26:0), Sphingomyelin-34:0 (SM 34:0), and Hexosylceramide-42:2 (HexCer 42:2), can be used as early warning biomarkers for damp syndrome for early warning of damp syndrome, that is, for use as biomarkers in the preparation of products for discriminating patients with damp syndrome among healthy populations.

[0008] Preferably, the product is a reagent, a kit, etc.

[0009] Preferably, the application is to discriminate patients with damp syndrome among healthy populations by quantitatively analyzing the relative abundances of LPC 16:0, LPC 22:0, PC 26:0, SM 34:0, and HexCer 42:2 in samples from healthy populations.

[0010] Preferably, the sample can be serum.

[0011] Preferably, the reagent is a targeted lipidomics analysis reagent.

[0012] The second object of the present invention is to provide the use of a reagent for detecting LPC 16:0, LPC 22:0, PC 26:0, SM 34:0, or HexCer 42:2 in the preparation of products for discriminating patients with damp syndrome among healthy populations.

[0013] Preferably, it is a reagent for detecting the relative abundances of LPC 16:0, LPC 22:0, PC 26:0, SM 34:0, or HexCer 42:2 in the serum of healthy populations.

[0014] The present invention discovers that five core lipids, namely LPC 16:0, LPC 22:0, PC 26:0, SM 34:0, and HexCer 42:2, all exhibit good area under the curve (AUC), accuracy, precision, sensitivity, and F1 recovery rate in the discovery cohort. This indicates that the five lipids, LPC 16:0, LPC 22:0, PC 26:0, SM 34:0, and HexCer 42:2, have the potential to become early warning biomarkers for damp syndrome. The present invention further uses an external validation cohort to evaluate the accuracy and precision of the five-core lipid classifier. As Figure 12 shown, the blue solid line indicates that the classifier constructed from these five core lipids has excellent performance in the discovery cohort (AUC = 0.914), and the red solid line indicates that the classifier constructed from these five core lipids also has excellent predictive ability in the external validation cohort (AUC = 0.935). In summary, the five core lipids, LPC16:0, LPC 22:0, PC 26:0, SM 34:0, and HexCer 42:2, can be used as early warning biomarkers for damp syndrome for early warning of damp syndrome. Description of the Drawings

[0015] Figure 1 is an overview diagram of the damp syndrome assessment scale for the included study population. (A) The proportion of Score 0 indicating none; the proportion of Score 1 indicating mild; the proportion of Score 2 indicating moderate; the proportion of Score 3 indicating severe; the proportion of Score 4 indicating extremely severe;

[0016] Figure 2 is the targeted lipidome database. (A) Lipid classes and their parent nuclear structures; (B) The number of lipid classes in the database;

[0017] Figure 3 is the number of significantly changed lipids in the damp syndrome population cohort. The dark blue module in the figure represents the number of lipids significantly changed in the damp syndrome body;

[0018] Figure 4 is the differential lipid expression profile. Red indicates upregulation; blue indicates downregulation; the total number of lipids is indicated inside the circle;

[0019] Figure 5 is the lipid co-network. The dots represent lipids; the color represents lipid classes; the symbols / symbols represent the direction and intensity of the correlation in the healthy population cohort / damp syndrome population cohort; the numbers represent the count of lipid pairs showing this specific change pattern in the lipid co-network;

[0020] Figure 6It is a network relationship diagram of seven lipids. The dots represent lipids; the colors of the dots represent lipid categories; the symbols / symbols represent the direction and strength of the correlation in the healthy population cohort / wet syndrome population cohort; the numbers represent the count of lipid pairs showing this specific change pattern in the lipid co - network;

[0021] Figure 7 It is the core lipid screening. The dots represent lipids, and the enlarged dots with text labels represent core lipids; the colors represent lipid categories; the symbols / symbols represent the direction and strength of the correlation in the healthy population cohort / wet syndrome population cohort; the numbers represent the count of lipid pairs showing this specific change pattern in the lipid co - network;

[0022] Figure 8 It is the correlation analysis of five core lipids and the total score of wet syndrome. The scatter plot plots the Spearman correlation between five core lipids and the total score of wet syndrome of all participants, and fits the regression line (red) and 95% confidence interval (gray area); *P < 0.05, **P < 0.01, ***P < 0.001;

[0023] Figure 9 It is that the core lipids are highly correlated with wet syndrome. (A) Heat map of the correlation analysis between core lipids and wet syndrome phenotype modules. (B) Wet syndrome phenotypes in four modules;

[0024] Figure 10 It is the technical roadmap of early warning biomarkers for wet syndrome;

[0025] Figure 11 It is the performance ability of five core lipids in the discovery cohort;

[0026] Figure 12 It is the ROC curve. The blue line fitting line represents the area under the curve in the discovery set; the red fitting line represents the area under the curve in the validation set. Detailed implementation mode

[0027] The following examples are for further illustration of the present invention rather than limitations thereof.

[0028] Example 1:

[0029] I. Experimental methods

[0030] 1.1 Clinical cohort recruitment

[0031] All clinical population cohorts involved in this article were uniformly recruited by Guangdong Provincial Hospital of Traditional Chinese Medicine. During this period, basic information such as the names, genders, ages, contact information, ID card information, and work units of the subjects was collected. Each subject participating in this trial was informed in detail about the relevant content of this scientific research, and the privacy information of the subjects was strictly protected. The subjects voluntarily decided to participate and signed the project informed consent form. This study was approved by the Institutional Ethics Committee of Guangdong Provincial Hospital of Traditional Chinese Medicine (BF2022-086-01). The cohort composition and sample size are as follows:

[0032] This invention consists of two data sets, a discovery set and a validation set. Each data set contains two clinical population cohorts: a healthy population and a damp syndrome population cohort. Among them, the discovery set contains 12 cases of the healthy population and 16 cases of the damp syndrome population; the validation set contains 23 cases of the healthy population and 22 cases of the damp syndrome population.

[0033] Inclusion criteria for the healthy population: ① The total score of damp syndrome ≤ 19 points; ② Voluntarily participate in and cooperate with this project research and sign the informed consent form.

[0034] Inclusion criteria for the damp syndrome cohort: The total score of damp syndrome ≥ 20 points, and it is judged as damp syndrome (when the total score of damp syndrome of the subject is between 20 and 39 points, it is judged as mild damp syndrome; the total score is between 40 and 59 points, it is judged as moderate damp syndrome; the total score of damp syndrome is between 60 and 79 points, it is judged as severe damp syndrome; the total score of damp syndrome ≥ 80 points, it is judged as extremely severe damp syndrome); Voluntarily participate in and cooperate with this project research and sign the informed consent form.

[0035] Exclusion criteria: ① Those with secondary blood glucose, blood lipid, and blood pressure abnormalities caused by related diseases such as liver and kidney functions (nephrotic syndrome, renal failure, etc.), and autoimmune deficiencies (systemic lupus erythematosus, etc.). ② Those with blood glucose and blood lipid abnormalities caused by drugs (such as phenothiazines). ③ Postoperative patients (malignant tumors, cardiovascular and cerebrovascular diseases, serious accidental traumas, etc.). ④ Pregnant, lactating, or planning to be pregnant women. ⑤ Those with allergic constitutions or those who cannot cooperate for a long time.

[0036] 2.1 Detection and analysis of biochemical index information

[0037] 2.1.1 Biochemical index information

[0038] The information collected for biochemical indexes is as follows:

[0039] Indicators such as height, weight, body mass index (BMI), waist circumference, hip circumference, waist-to-hip ratio (WHR), systolic blood pressure (SBP), creatinine (CR), total cholesterol (TC), uric acid (UA), aspartate transaminase (AST), fasting plasma glucose (GLU), high density lipoprotein cholesterol (HDLC), fasting serum insulin (INS), insulin resistant index (HOMA-IR), C-reactive protein (CRP), tumor necrosis factors (TNF), interleukin-6 (IL6), total triglyceride (TG), white blood cell (WBC), hemoglobin (HB), red blood cell (RBC), alanine transaminase (ALT), urea nitrogen (UREA), blood platelet (PLT), diastolic blood pressure (DBP), low density lipoprotein cholesterol (LDLC). Before blood collection, subjects were required to fast.

[0040] 2.1.2 Clinical serum sample collection

[0041] The clinical serum sample collection method is as follows:

[0042] (1) Use a 5 mL blood collection tube to collect 4 mL of forearm venous blood from each subject for whole blood samples;

[0043] (2) Then place the blood collection tube in a centrifuge and centrifuge at 4°C for 15 minutes, and take the supernatant (3000 rcf) to obtain the serum sample;

[0044] (3) To prevent repeated freezing and thawing of the samples, finally, the serum samples of each subject were aliquoted into 1.5 mL centrifuge tubes (100 μL per tube) and stored at -80 °C in a refrigerator for cryopreservation, awaiting subsequent analysis.

[0045] 2.1.3 Detection of biochemical index information

[0046] The above-mentioned clinical biochemical index detections were all completed at the Research Center for Health State Identification and Preventive Treatment of Disease with Traditional Chinese Medicine in Guangdong Provincial Hospital of Traditional Chinese Medicine.

[0047] 2.1.4 Data statistical analysis

[0048] The steps of data statistical analysis are as follows:

[0049] The results of biochemical index data were presented in the form of mean ± standard deviation (Mean ± SD). For data conforming to the normal distribution, a t-test was used between two groups, and one-way analysis of variance (one-way ANOVA) was used among multiple groups. For non-normal data analysis, the Wilcoxon rank sum test was used between two groups, and the Kruskal-Wallis test was used among multiple groups; a paired t-test was used within groups conforming to the normal distribution. A paired Wilcoxon rank sum test was used within groups of non-normal data. When P < 0.05, it was considered to have statistical significance, and all statistical analyses were performed using R software.

[0050] 3.1 Information on the dampness syndrome assessment scale

[0051] Information on the traditional Chinese medicine dampness syndrome assessment scale is as follows:

[0052] Sticky tongue fur, watering mouth with saliva, thick tongue fur, heavy headness as if swathed, feeling cumbersome and heavy, fixed heavy limbs, soreness of joints, sticky mucous stool, sticky and greasy in mouth, abnormal leukorrhea secretion / oily scrotum, oily skin with greasy hair, prone obese figure, unsurfaced fever, sticky and sweaty, fatigue, sleepy, dim complexion, sticky eye discharge, bad breath, thirsty with no desire for fluids, phlegm hypersecretion, a bloating stomach, poor appetite, nausea and vomiting, feels cumbersome and heavy in waist, shapeless pasty stool, increased defecate frequency, skin ulcer, tongue with two saliva foaming lines, enlarged tongue. There are 30 phenotypes in total, and each phenotype is divided into five grades: none, mild, moderate, severe, and extremely severe, corresponding to 0 points, 1 point, 2 points, 3 points, and 4 points respectively. The full score is 120 points. The higher the score, the higher the severity of damp syndrome.

[0053] 4.1 Targeted lipidomics analysis

[0054] 4.1.1 Serum sample preparation

[0055] The preparation steps of the targeted lipidomics serum samples are as follows:

[0056] (1) Take 100 μL of clinical serum samples, add 4 μL of deuterium-labeled lipid mass spectrometry mixed standard, and then add 400 μL of pre-cooled methyl tert-butyl ether and 80 μL of methanol solution (MTBE:MeOH:H2O =

[0057] 10:2:2.5, v / v);

[0058] (2) Vortex for 30 s and centrifuge at 3000 rcf at 4 °C for 15 min;

[0059] (3) Take 300 μL of the supernatant and transfer it to a new 1.5 mL centrifuge tube, and freeze-dry;

[0060] (4) Add 100 μL of pre-cooled dichloromethane:methanol solution (1:1, v / v), vortex for 30 s to dissolve it completely;

[0061] (5) Take 5 μL of each sample and mix well as the QC sample, and detect the QC sample every 8 samples;

[0062] (6) Conduct serum targeted lipidomics research using UHPLC-Q-TRAP-MS / MS 6500plus.

[0063] 4.1.2 Liquid chromatography-mass spectrometry detection conditions

[0064] Liquid chromatography conditions: Use a Waters (USA) reversed-phase BEH C8 chromatographic column (2.1×100 mm, 1.8 μm), the column oven temperature is 55 °C, mobile phase: A is ACN:H 2 O (60:40, v / v), B is IPA:CAN (90:10, v / v), and the gradient elution program is shown in Table 1.

[0065] Table 1 Gradient elution program

[0066]

[0067] Use the AB Sciex 6500plus mass spectrometry system for lipid detection. The optimized parameters are as follows: ESI temperature: 550 °C, ISVF are 5500 V (+), -4500 V (-) respectively, nebulizing gas: 55 psi, CUR: 35 psi, multiple reaction monitoring mode, and the data acquisition and analysis use OS-Q software (version 1.4.1.20719, AB Sciex, USA).

[0068] 5.1 Data analysis

[0069] The steps of targeted lipidomics analysis are as follows:

[0070] (1) Peak table extraction: Use Sciex OS software to process the off-machine data of targeted lipidomics. Click on the Analytics module, set the path where the original files are located, select any one sample file as the quantitative analysis method template, click on Workflow, after selecting and importing all wiff files, click on the Integration module and set the peak integration parameters as shown in Table 2 below, and save the quantitative method. Then click on the New option under Results, set the path of the original files, select and import all wiff files, select the quantitative analysis method template file established previously, click on Process, and after the analysis is completed, the lipidomics peak table data can be obtained.

[0071] Table 2 Optimized parameters for lipid peak table extraction

[0072]

[0073] (2) Screening of differential lipids: Use the built-in "Statistical Analysis" function in MetaboAnalyst to filter and fill the missing values in the peak table matrix. The filtering method deletes the observations with missing values greater than 50%, and the filling method selects the K-Nearest Neighbor (KNN) algorithm. In order to make the data comparable, and unify the response intensity sizes of all variables of the same sample on the same standard, and improve the normal distribution of the data to reduce errors, after normalizing the data (group PQN: QC), performing a logarithmic transformation (Log transformation: base 10), and scaling (Pareto scaling), screen for differential metabolites. The screening principle is compounds with a differential fold change (FC) ≥ 2.0 and P < 0.05 or FC ≤ 0.05 and P < 0.05. Use multivariate statistical analysis to observe the discrimination between groups, and use the ropls package (version 1.34.0) for principal component analysis (Principal Components Analysis, PCA analysis) for unsupervised dimensionality reduction of the data, and partial least squares discriminant analysis (Partial least squares analysis, PLS-DA analysis) for supervised dimensionality reduction of the data.

[0074] (3) Lipid network construction: Use the DGCA package (version 1.0.3) to calculate the relationships between lipids in the healthy and damp syndrome states, and identify abnormal lipid interaction relationships caused by the occurrence of damp syndrome. The specific steps are as follows: After importing the lipid peak list data matrix, use the ddcorAll function to run a complete differential gene correlation analysis, compare the spearman correlations in the healthy and damp syndrome states, and only include lipid pairs with significant differences (P<0.05) for analysis. Then, use the MEGENA package (version 1.3.7) to establish a correlation network of lipid pairs between the healthy cohort and the damp syndrome cohort to reveal the lipid co-regulation network and screen for core lipids. The specific steps are as follows: First, import the data matrix obtained from the DCGA analysis, and use the calculate.PFN function in the package to calculate the edge pair sorted list; then, use the do.MEGENA function to perform multi-scale embedded co-expression network analysis; finally, after the analysis is completed, use the plot_module function to draw the network interaction module, from which hublipids can be screened. Data visualization is performed using cytoscape (https: / / cytoscape.org / ) software for network adjustment and drawing.

[0075] (4) Module analysis: Use the WGCNA package (version 1.72-1) to perform modular processing on the damp syndrome phenotype and biochemical indicators. The specific steps are as follows: After importing the data, first, use the goodSamplesGenes function in the package to check for missing values and identify outliers. When the output result is TRUE, it indicates that there are no missing values; then use the pickSoftThreshold function in the package to screen for the optimal soft threshold. The screening principle is to select the first value that reaches above 0.6, which is the soft threshold for this analysis, and examine the scalability of the above soft threshold. After confirming the good soft threshold, next, use the blockwiseModules function in the package to construct an automated network in one step. After the analysis is completed, use the plotDendroAndColors function to draw the hierarchical clustering map of module labels. Finally, perform an analysis of the module-phenotype data association. During this analysis process, perform a correlation analysis between the module and the phenotype to find important associations between the two, save the correlation results locally, and use the ggplot2 package for visual analysis. Use the ggpmisc package to draw the spearman correlation graph between the core lipids and the damp syndrome phenotype, and add the linear regression equation and P value.

[0076] (5) Machine learning: The earth package (version 5.3.2) was used to perform multivariate adaptive regression spline regression analysis to screen potential biomarkers for early prediction of wet syndrome. The specific steps are as follows: First, the discovery set and validation set data were imported; then, the earth function was used to build a machine learning model classifier in the discovery set data, where "pmethod" selected "cv" cross validation, "ncross" was set to 3, and "glm" selected the logistic regression model binomial and ran. After the model was built, it was applied to the validation set data to calculate the performance. The roc_aucha function in the yardstick package (version 1.2.0) was used to calculate the AUC values ​​in the discovery set and validation set, respectively, and the roc_curve function was used in combination with the ggplot2 package to draw the ROC curve.

[0077] 2. Experimental Results

[0078] The cohort included in this study was a healthy population based on Western medical indicators. Figure 1 As shown in Figure 1, dampness syndrome covers 30 phenotypic information in TCM clinical practice, involving mental state, oral cavity, feces, metabolism, etc. People with a total score of dampness syndrome higher than 20 points are considered to have dampness syndrome (DS).

[0079] Based on the two key indicators of incidence and severity, 10 characteristic phenotypes of dampness syndrome were screened out with incidence ≥ 80% and severity ≥ 25% respectively: oily skin with greasy hair, dim complexion, fatigue, sleepiness, fixed heavy limbs, tongue with two salivafoaming lines, thick tongue fur, sticky and greasy in mouth, sticky mucous stool, and shapeless pasty stool. The incidence and severity of the 10 characteristic phenotypes are shown in Table 1. Figure 1 The above results suggest that, on the one hand, people with dampness syndrome may be accompanied by abnormal metabolism (greasy face or hair, dirty complexion, fatigue, drowsiness / sleepy); on the other hand, dampness syndrome may cause disturbances in oral and intestinal microorganisms. Based on the results of the above characteristic phenotypic screening, we studied the lipid metabolism in people with dampness syndrome from the perspective of lipid metabolism in the body.

[0080] According to the lipid classification system proposed by the "Lipid Metabolism Pathway Research Program" (LIPID MAPS) of the US NIH in 2003, lipids are divided into 8 major categories. Among them, the blood lipid categories widely present in human plasma are 8 major categories: Fatty acids, Sphingolipids, Glycerolipids, Glycerophospholipids, Cholesterol ester, Prenol lipids, Polyketides, and Saccharolipids. As Figure 2 shown in A, this method includes >17 lipid categories. As Figure 2 shown in B, this method covers more than 1000+ lipid ion pair information, including fatty acid classes (FA), diacylglycerol classes (DAG), triacylglycerol classes (TAG), cholesterol ester classes (CE), phosphatidylethanolamine classes (PE), lysophosphatidylethanolamine classes (LPE), phosphatidylcholine classes (PC), lysophosphatidylcholine classes (LPC), phosphatidylinositol classes (PI), lysophosphatidylinositol classes (LPI), phosphatidic acid classes (PA), lysophosphatidic acid classes (LPA), phosphatidylglycerol classes (PG), lysophosphatidylglycerol classes (LPG), ceramide classes (Cer), sphingomyelin classes (SM), hexosylceramide classes (HexCer) and other lipids. Among them, DAG: 260, TAG: 192, Cer: 150, HexCer: 88, SM: 48, PC: 53, LPC: 20, PI: 77, LPI: 19, PE: 67, LPE: 23, PG: 67, LPG: 19, PA: 51, LPA: 19, PS: 72, LPS: 19, FA: 25, CE: 22, which provides technical support for the development of targeted lipidomics detection.

[0081] Inter-group differences are usually reflected in the content changes of differential metabolites. These differential changes involve multiple metabolic pathways and biological processes, which not only expand the understanding of the in vivo metabolic network of the damp syndrome population cohort, but also provide key clues for further research on the pathological state of damp syndrome. As Figure 3 shown, after normalizing, standardizing, and scaling the data, based on the screening principle of differential lipids: FC≥2 and P<0.05 or FC≤0.5 and P<0.05, differential lipids under each lipid category are screened in the damp syndrome population cohort. The PC class changes the most, followed by the Cer class, DAG class, LPC class, etc.

[0082] To explore the disorder and perturbation levels of differential lipids during the process of damp syndrome, we further constructed the differential lipid expression profiles of the damp syndrome population. As Figure 4 shown, we observed large-scale fluctuations in the number of lipids in the lipid conversion network caused by damp syndrome. It was mainly reflected in two aspects. One was the conversion of lysophosphatidylcholine (LPC) to phosphatidylcholine (PC), and the other was the interaction between ceramide (Cer) and sphingomyelin (SM). The above results indicated that damp syndrome interfered with glycerophospholipid metabolism and sphingolipid metabolism, which was consistent with the previous research findings.

[0083] Co-regulated genes usually exhibit similar gene expression patterns, which means that their gene expression levels are strongly correlated. Similarly, under the same circumstances, the strong correlation between lipid levels indicates that these lipids may coexist and regulate each other along the same metabolic pathway. Once the occurrence of a disease disrupts their mutual regulatory effects, it indicates the occurrence of disease-related metabolic disorders. Next, we used differential gene co-expression analysis (DGCA) to identify the lipid pair relationships that were abnormal due to the interference of damp syndrome and constructed their co-expression network. As Figure 5 shown, compared with the healthy population cohort, a total of 809 lipid pair interaction relationships changed significantly in the damp syndrome population cohort.

[0084] As Figure 6As shown in the figure, in the state of dampness syndrome, seven pairs of lipid relationships appear in the body: + / +, - / 0, + / 0, 0 / -, 0 / +, + / -, and - / +, indicating that dampness syndrome disrupts the normal lipid interaction network in the body. + / + (red connecting line) indicates that 14 lipid pairs show a positive correlation in the healthy population cohort and maintain a positive correlation in the dampness syndrome population cohort, indicating that the normal functions of these 14 lipid pairs are not affected in the state of dampness syndrome. - / 0 (green connecting line) indicates that 181 lipid pairs show a negative correlation in the healthy population cohort but are uncorrelated in the dampness syndrome population cohort, indicating that in the state of dampness syndrome, the lipid pairs that originally showed a negative correlation in the healthy population cohort are inhibited. + / 0 (orange connecting line) indicates that 295 lipid pairs show a positive correlation in the healthy population cohort but are uncorrelated in the dampness syndrome population cohort, indicating that in the state of dampness syndrome, the lipid pairs that originally showed a positive correlation in the healthy population cohort are inhibited. 0 / - (brown connecting line) indicates that 98 lipid pairs are uncorrelated in the healthy population cohort but show a negative correlation in the state of dampness syndrome, indicating that in the state of dampness syndrome, the lipid pairs that originally showed an uncorrelated relationship in the healthy population cohort show a negative correlation. 0 / + (pink connecting line) indicates that 120 lipid pairs are uncorrelated in the dampness syndrome population cohort but show a positive correlation in the state of dampness syndrome, indicating that in the state of dampness syndrome, the lipid pairs that originally showed an uncorrelated relationship in the healthy population cohort show a negative correlation. + / - (cyan connecting line) indicates that 55 lipid pairs show a positive correlation in the healthy population cohort but a negative correlation in the dampness syndrome population cohort, indicating that in the state of dampness syndrome, the lipid pairs that originally showed a positive correlation in the healthy population cohort are reversed. - / + (blue connecting line) indicates that 46 lipid pairs show a negative correlation in the healthy population cohort but a positive correlation in the dampness syndrome population cohort, indicating that in the state of dampness syndrome, the lipid pairs that originally showed a negative correlation in the healthy population cohort are reversed. The above results indicate that during the occurrence and development of dampness syndrome, it greatly affects lipid interaction relationships, thereby affecting their biological functions and potentially causing irreversible damage.

[0085] By constructing a lipid interaction network and identifying co-regulated lipid pairs, it helps to reveal the potential functional modules and lipid regulatory networks affected by dampness syndrome. In the lipid interaction network, individual lipids act as key nodes in the network, participating in multiple lipid pair transformation axes and regulatory networks to maintain the stability of the regulatory network. Therefore, screening for core lipids in the abnormal lipid network in the state of dampness syndrome is of great significance for understanding the potential pathogenesis of dampness syndrome and timely warning and intervention. As Figure 7 shown, in the state of dampness syndrome, five core lipids play a key role in the interaction network, namely PC 26:0, LPC 16:0, LPC 22:0, SM 34:0, and HexCer 42:2, including three phosphatidylcholine classes and two sphingolipid lipids.

[0086] Spearman correlation analysis (Spearman analysis) was performed on the five core lipids and the total score of the dampness syndrome assessment scale. As Figure 8 shown, the five core lipids, LPC 22:0, LPC 16:0, HexCer 42:2, PC 26:0, and SM 34:0, were correlated with the total score of the dampness syndrome to varying degrees. Among them, LPC 22:0, LPC 16:0, PC 26:0, and HexCer 42:2 were significantly negatively correlated with the total score of the dampness syndrome. The above results suggest that the five core lipids may be involved in the occurrence and development of the dampness syndrome and may be biomarkers for early warning of the dampness syndrome.

[0087] Thirty dampness syndrome-related phenotypes were modularized. Its advantage is that it can modularize and cluster co-synergistic phenotypes, thereby removing redundant information and reducing the mutual interference between individual phenotypes. As Figure 9 shown, modular clustering analysis of the dampness syndrome phenotypes yielded four modules, namely Module I (M1), Module II (M2), Module III (M3), and Module IV (M4). The results of the correlation analysis between the core lipids and the dampness syndrome phenotype modules are as Figure 9 shown in B. The five core lipids, LPC 16:0, LPC 22:0, PC 26:0, SM34:0, and HexCer 42:2, were correlated with the four modules to varying degrees. Among them, HexCer 42:2 was significantly negatively correlated with M1, M2, and M3, indicating that the core lipid HexCer 42:2 may be involved in the phenotypic evolution of the three modules M1, M2, and M3. The three core lipids, LPC 16:0, LPC 22:0, and PC 26:0, were simultaneously significantly correlated with the four modules M1, M2, M3, and M4, indicating that the three lipids may play a co-synergistic role in the phenotypes covered by M1, M2, M3, and M4. However, SM34:0 was not significantly correlated with the four core lipids, suggesting that SM 34:0 may interact with other lipids to affect the correlation of the dampness syndrome phenotype modules. The phenotypic composition in each module is as Figure 9As shown in Figure B, the phenotypes included in the module have certain characteristics. For example, the M1 module contains 7 damp syndrome phenotypes, namely Tongue with twosaliva foaming lines, Sticky and greasy in mouth, Wateringmouth with saliva, Sticky tongue fur, Bad breath, Increased defecate frequency, and A bloating stomach, mainly involving phenotypes related to the oral cavity; the M2 module contains 6 damp syndrome phenotypes, such as Poor appetite, Sticky eyedischarge, Sleepy, Fatigue, Feels cumbersome andheavy, and Heavy headness as if swathed, mainly involving manifestations such as mental state; the M3 module contains 4 damp syndrome phenotypes, namely Shapeless pasty stool, Sticky mucous stool, Feels cumbersome and heavy in waist, and Thicktongue fur, involving characteristics related to the intestine; the M4 module contains 13 damp syndrome phenotypes, namely Thirsty with no desire for fluids, Skin ulcer, Unsurfaced fever, Phlegm hypersecretion, Prone obese figure, Nausea and vomiting, Abnormal leukorrheasecretion, Soreness of joints, Sticky andsweaty, Fixed heavy limbs, Oily skin with greasyhair, Dim complexion, and Enlarged tongue, involving phenotypes related to metabolism and endocrinology. The above clustering results are relatively consistent with the previous clustering results of the screened characteristic phenotypes of damp syndrome.

[0088] The five core lipids were significantly correlated with the total score of dampness syndrome, the dampness syndrome phenotype module, and the characteristic phenotypes of dampness syndrome. As early warning biomarkers, they not only need to show excellent performance in the discovery cohort, but also be able to demonstrate stable robustness and predictive ability in the validation cohort. Based on this, we first evaluated the performance of the five core lipids as early warning biomarkers for dampness syndrome in the discovery set. As Figure 10 shown, the technical roadmap for validating early warning biomarkers for dampness syndrome is presented.

[0089] As Figure 11 shown, the five core lipids LPC 16:0, LPC 22:0, PC 26:0, SM 34:0, and HexCer all showed good area under the curve (AUC), accuracy, precision, sensitivity, and F1 recovery rate in the discovery cohort. This indicates that the five lipids, LPC 16:0, LPC 22:0, PC 26:0, SM 34:0, and HexCer, have the potential to be early warning biomarkers for dampness syndrome.

[0090] A machine learning classifier was established based on the above five core lipids, and an external validation cohort was used to evaluate the accuracy and precision of the five-core lipid classifier. As Figure 12 shown, the blue solid line indicates that the classifier constructed by these five core lipids has excellent performance in the discovery cohort (AUC = 0.914), and the red solid line indicates that the classifier constructed by these five core lipids also has excellent predictive ability in the external validation cohort (AUC = 0.935). In summary, the five core lipids LPC 16:0, LPC 22:0, PC 26:0, SM 34:0, and HexCer 42:2 can be used as early warning biomarkers for dampness syndrome for early warning of dampness syndrome.

Claims

1. Use of lipids Lysophosphatidyl choline-16:0, Lysophosphatidyl choline-22:0, Phosphatidylcholine-26:0, Sphingomyelin-34:0 and / or Hexosylceramide-42:2 as early warning biomarkers of dampness syndrome in the preparation of products for distinguishing patients with dampness syndrome in healthy people.

2. The use according to claim 1, characterized in that: The products described are reagents and test kits.

3. The use according to claim 1, characterized in that: The application is to identify patients with dampness syndrome in the healthy population by quantitatively analyzing the relative abundance of Lysophosphatidyl choline-16:0, Lysophosphatidyl choline-22:0, Phosphatidylcholine-26:0, Sphingomyelin-34:0, and Hexosylceramide-42:2 in samples of the healthy population.

4. The use according to claim 1, characterized in that: The sample is serum.

5. The use according to claim 2, characterized in that: The reagent is a targeted lipidomics analysis reagent.

6. Use of reagents for detecting Lysophosphatidyl choline-16:0, Lysophosphatidyl choline-22:0, Phosphatidylcholine-26:0, Sphingomyelin-34:0 and / or Hexosylceramide-42:2 in the preparation of products for distinguishing patients with dampness syndrome in healthy people.

7. The use according to claim 6, characterized in that A reagent for detecting the relative abundance of Lysophosphatidylcholine-16:0, Lysophosphatidyl choline-22:0, Phosphatidylcholine-26:0, Sphingomyelin-34:0 and / or Hexosylceramide-42:2 in the serum of healthy people.

Images