Application of reagent for detecting REN protein in preparation of product for evaluating myocardial infarction risk of subject in future
By detecting the expression amount of REN protein and the enzyme-linked immunization kit, combining machine learning models to evaluate the REN protein concentration, the problem of insufficient accuracy of the risk assessment of central myoinfarction in the prior art was solved, and high accuracy and economic risk prediction were achieved.
Patent Information
- Application Number
- CN202411165332.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-23
- Publication Date
- 2024-11-12
AI Technical Summary
The existing myocardial infarction risk assessment tools rely on traditional risk factors and cannot accurately predict future myocardial infarction events, and cannot consider the complex pathophysiological processes behind myocardial infarction.
By detecting the expression of REN protein, the REN protein concentration in the subject's peripheral blood is evaluated using enzyme-linked immune kits and machine learning models (such as lightGBM), and converted to NPX values to determine whether the subject will experience myocardial infarction in the next 10 years. risk.
It has achieved early accurate prediction of myocardial infarction risk, with high prediction accuracy, non-invasiveness, low economic cost, simplicity, and the detection results exceed the prediction level of traditional methods.
Smart Images

Figure CN118937699A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the field of biomedical detection technology, and in particular relates to the use of a reagent for detecting REN protein in the preparation of a product for assessing the risk of a subject suffering from myocardial infarction in the future. Background Art
[0002] Cardiovascular disease, especially ischemic heart disease, is one of the leading causes of death worldwide and significantly increases the risk of disability. Despite improvements in reperfusion therapy and prognosis, myocardial infarction still leads to various complications, including cardiogenic shock and heart failure, which illustrates the importance of identifying risk factors for early prevention. Current risk assessment tools, such as the Framingham Risk Score and the Systematic Coronary Artery Risk Evaluation (SCORE2) system, have the potential to estimate the risk of developing coronary heart disease within 10 years. However, these tools rely solely on traditional risk factors and are limited in accurately predicting future myocardial infarction events due to the complex interactions of factors such as blood pressure, smoking, and cholesterol levels. In addition, these tools may not take into account the intricate and shared pathophysiological processes behind myocardial infarction.
[0003] Renin (REN) is a proteolytic enzyme released by the juxtaglomerular cells of the juxtaglomerular apparatus (also known as the juxtaglomerular complex). The main function of renin is to maintain renal filtration pressure by ultimately increasing blood pressure: renin produces angiotensin I by cleaving angiotensinogen synthesized in the liver. Angiotensin I can be further cleaved into angiotensin II by angiotensin converting enzyme (ACE), thereby efficiently constricting blood vessels, increasing the secretion of aldosterone and antidiuretic hormone, stimulating the hypothalamus to produce thirst, and ultimately increasing blood pressure. It has been reported in the past that REN has the effects of regulating the body's water and sodium balance, lowering blood pressure, and reducing the osmotic pressure of extracellular fluid, but there are still no reports on the use of REN alone to predict the long-term risk of myocardial infarction. Summary of the invention
[0004] In view of this, the object of the present invention is to provide a reagent for detecting REN protein for use in the preparation of a product for assessing the risk of myocardial infarction in a subject in the future.
[0005] In order to achieve the above-mentioned object of the invention, the present invention provides the following technical solutions:
[0006] The present invention provides an application of a reagent for detecting REN protein in preparing a product for assessing the risk of a subject suffering from myocardial infarction in the future.
[0007] Preferably, the reagent for detecting REN protein includes a reagent for detecting the expression level of REN protein.
[0008] Preferably, the reagent comprises an antibody against the REN protein.
[0009] Preferably, the reagent comprises an enzyme-linked immunosorbent assay kit.
[0010] Preferably, the future time is more than 10 years.
[0011] Preferably, the sample from the subject comprises peripheral blood.
[0012] Preferably, the product comprises a reagent or a kit.
[0013] Preferably, the evaluation standard is that when the NPX value of REN is ≥ 0.476, the risk of the subject suffering from myocardial infarction in the future is high, and when the NPX value of REN is < 0.476, the risk of the subject suffering from myocardial infarction in the future is low.
[0014] The present invention provides a system for assessing a subject's future risk of myocardial infarction, comprising a detection module and a judgment module, wherein the detection module is used to obtain the REN protein concentration in plasma and convert the REN protein concentration into a REN protein NPX value, and the judgment module comprises a readable carrier storing a judgment standard.
[0015] Preferably, the judgment standard is that when the NPX value of REN is ≥ 0.476, the risk of the subject suffering from myocardial infarction in the future is high, and when the NPX value of REN is < 0.476, the risk of the subject suffering from myocardial infarction in the future is low.
[0016] Compared with the prior art, the present invention has the following beneficial effects:
[0017] The present invention provides an application of a reagent for detecting REN protein in the preparation of a product for assessing the risk of myocardial infarction in the future of a subject. The present invention can accurately predict / assess the risk of myocardial infarction in the subject in the next 10 years or more by detecting the concentration level of REN protein in peripheral blood, and provide early risk guidance for people who need intervention. The application of the present invention has a high prediction accuracy rate, and is simple, non-invasive, and has low economic cost. It only needs to detect the level of REN in the peripheral micro-blood of the subject, and does not require the clinical indicator level required in the traditional model. The detection accuracy also exceeds the prediction level of the traditional method, and has a high application prospect. BRIEF DESCRIPTION OF THE DRAWINGS
[0018] Figure 1 This is a protein spectrum with a high risk of myocardial infarction screened using the Cox risk regression model;
[0019] Figure 2 ROC curve of the lightGBM machine learning model built for the REN protein;
[0020] Figure 3 Kaplan-Meier survival analysis was used to obtain the survival rate results of samples in the high-risk group and the low-risk group, where the green line represents the low-risk group and the red line represents the high-risk group. DETAILED DESCRIPTION
[0021] The present invention analyzes the results of the highly sensitive Olink proteomics of the world's largest natural population cohort to date (a total of 2,923 proteins in more than 50,000 people were measured, with a follow-up period of about 14 years), and uses the highly efficient lightGBM machine learning model to mine plasma protein molecular markers, thereby maximizing the rational use of existing data, thereby achieving a high prediction accuracy for the occurrence of myocardial infarction (such as ST-segment elevation myocardial infarction and non-ST-segment elevation myocardial infarction) in the next 10 years or more.
[0022] In the early stage of screening protein markers, the present invention incorporates different personal background characteristics, clinical characteristics or proteomic data for machine learning testing to balance the prediction accuracy and the number of included indicators. The Cox risk regression model was used to screen the protein spectrum related to myocardial infarction from 2923 proteins, and then the machine learning algorithm lightGBM was used to rank the importance of proteins in the prediction of myocardial infarction. REN was a high-ranking protein, and the lightGMB machine learning model constructed by the single REN protein had a higher prediction performance than the model constructed by traditional clinical indicators.
[0023] Based on this, the present invention provides an application of a reagent for detecting REN protein in the preparation of a product for assessing a subject's future risk of myocardial infarction.
[0024] In the present invention, the reagent for detecting REN protein preferably includes a reagent for detecting the expression of REN protein, and further preferably includes an antibody for REN protein, and further preferably includes an enzyme-linked immunosorbent assay kit. The enzyme-linked immunosorbent assay kit is preferably an Olink kit, which is purchased from Thermo Fisher Scientific, product name Explore 3072 / 384. The present invention can be used to assess the risk of myocardial infarction in the future of a subject by detecting the concentration of REN protein, wherein the future time is more than 10 years, such as 10 years, 11 years, 12 years, 13 years, 14 years or 15 years.
[0025] In the present invention, the sample of the subject includes peripheral blood, and the REN protein concentration in the plasma is detected to assess the risk of myocardial infarction in the subject in the future, which is a non-invasive, simple and low-cost method. The product preferably includes a reagent or a kit. The evaluation standard is that when the NPX value of REN is ≥0.476, the risk of myocardial infarction in the subject in the future is high, and when the NPX value of REN is <0.476, the risk of myocardial infarction in the subject in the future is low.
[0026] The present invention provides a system for assessing a subject's future risk of myocardial infarction, comprising a detection module and a judgment module, wherein the detection module is used to obtain the REN protein concentration in plasma and convert the REN protein concentration into a REN protein NPX value, and the judgment module comprises a readable carrier storing a judgment standard.
[0027] In the present invention, the judgment standard is that when the NPX value of REN is ≥ 0.476, the risk of the subject suffering from myocardial infarction in the future is high, and when the NPX value of REN is < 0.476, the risk of the subject suffering from myocardial infarction in the future is low.
[0028] In the present invention, unless otherwise specified, all raw material components are commercially available products well known to those skilled in the art.
[0029] The technical solutions provided by the present invention are described in detail below in conjunction with the embodiments, but they should not be construed as limiting the protection scope of the present invention.
[0030] In the following embodiments, the present invention has no particular limitation on the Olink Explore proximity extension analysis method, and any method known in the art may be used, such as the method described in document DOI10.1038 / s41586-023-06592-6.
[0031] The present invention uses an Olink kit to measure protein. After obtaining the sample protein level, the sample protein level is converted into a normalized protein expression (NPX) value. The present invention has no particular limitation on the calculation method of the NPX value, and a method known in the art can be used, such as the method described in https: / / biobank.ndph.ox.ac.uk / show case / ukb / docs / Olink_1536_B0_to_B7_Normalization.pdf (UKB-Olink Explore 1536-Data NormalizationStrategy).
[0032] Example 1
[0033] Clinical data, basic information, and omics data of 51,294 subjects from UKBiobank with Olink proteomics measurement data were analyzed.
[0034] (1) Peripheral blood samples collected using EDTA tubes were centrifuged at 2500 g for 10 min at 4°C to separate plasma, and the supernatant (plasma) was aliquoted and stored at -80°C.
[0035] (2) The obtained plasma samples were subjected to protein quantification using the Olink Explore proximity extension assay, i.e., 2923 unique proteins were detected using the Olink kit to obtain the protein levels of 2923 samples, which were then converted into standardized protein expression (NPX) values.
[0036] (3) The NPX values of 2923 proteins were used to construct a Cox risk regression model, where the outcome of the Cox risk regression model was a binary outcome: whether myocardial infarction occurred, and the variables were the NPX values of the 2923 proteins + a series of calibrated covariates (sex, age, education level, body mass index (BMI), high-density lipoprotein (HDL), low-density lipoprotein (LDL), smoking, total triglycerides (TG), total cholesterol (TC), systolic blood pressure (SBP), diastolic blood pressure (DBG), estimated glomerular filtration rate (eGFR), whether diabetes occurred, and C-reactive protein level). Cox risk regression model = coxph [surv_obj (annual survival rate of surviving subjects) ~ group (risk group) + age (age) + sex (sex) + edu (years of education), data = mydf (Olink proteomics dataset)], after the regression risk model was constructed, the coef (cox_model) function was used to obtain the risk coefficient of the risk model, which is the risk value. Cox risk regression was performed on each of the 2923 proteins to obtain the p-value of each single protein. The p-value was then corrected for multiple tests and proteins with p<0.05 were screened to obtain a protein spectrum with a high risk of myocardial infarction.
[0037] Figure 1 The results showed that 270 proteins were identified in the protein profile with a high risk of myocardial infarction and were associated with myocardial infarction outcomes, among which the risk value (HR) of REN was 1.171 and the p value was 3.20×10 -7 .
[0038] (4) The protein spectra associated with a high risk of myocardial infarction were incorporated into the lightGBM machine learning model one by one, and the accuracy of the lightGBM machine learning model was evaluated by protein pair analysis.
[0039] lightGBM is an efficient machine learning algorithm based on the gradient boosting framework. It uses the gradient boosting framework to iteratively add decision trees to approximate the optimal solution of the objective function. By gradually and iteratively adding protein information, it continuously evaluates whether the model has achieved the optimal prediction effect.
[0040] Specifically, lightgbm (version 4.3.0) is used for model prediction. The model calculates the gradient (first-order derivative) and Hessian matrix (second-order derivative) of the loss function to find the optimal tree structure. For a given loss function L, the gradient g and Hessian h are:
[0041] L(y, Y) = (yY) 2
[0042]
[0043] Among them, y is the true value, Y is the predicted value, F(x i ) represents the predicted value of the current model on sample i.
[0044] After the above optimization, the lightGBM machine learning model is configured according to the following key parameters: specify the evaluation target as binary classification (objective = "binary"), the evaluation standard as AUC (metric = "auc"), perform data imbalance processing (is_unbalance = TRUE), the learning rate is 0.01 (learning_rate = 0.01), the maximum depth of the tree is 15 (max_depth = 15), the number of leaves is 10 (num_leaves = 10), the subsampling ratio of the data is 0.7 (subsample = 0.7), the subsampling ratio of the feature is 0.7 (colsample_bytree = 0.7), and the number of training rounds is 500 (nrounds = 500).
[0045] Next, the "Leave-One-Out" method was used for cross-validation: 51,294 subjects were classified into 10 regional categories, and an iteration was set for each region, in which the data of the region was used as a test set, and the data of the remaining nine regions were combined as a training set. The data consists of two parts: outcome variables and predictor variables. The outcome variable is whether myocardial infarction occurs, and the predictor variable is proteomic data. This will be repeated 10 times, and a different region will be selected as the test set each time. After all 10 iterations are completed, the performance of the model on the test set of each region is summarized, and the average performance index is calculated to establish the optimal LightGBM machine learning model.
[0046] The lightGBM machine learning model was used to dynamically obtain the prediction and test data of 51,294 cases of protein group blood. The prediction data refers to the simulated prediction of whether each individual has developed a myocardial infarction using the constructed lightGBM model and protein indicators; the test data refers to whether each individual has actually developed a myocardial infarction. The pROC (version 1.18.5) encapsulated in the R language program was used to calculate the receiver operating characteristic (ROC) and the area under the curve (AUC). Finally, a ROC curve evaluation model containing the true positive rate (TPR) and the false positive rate (FPR) was created to evaluate the average performance index. The specific steps are (where y is the true value and Y is the predicted value): ROC = roc(y, Y); AUC = auc(roc_res); TPR = roc_res$sensitivities; FPR = 1-roc_res$specificities.
[0047] Youden index is a statistical method used to evaluate the performance of a classifier. It can find the optimal classification threshold by maximizing the sum of sensitivity and specificity. The coords function encapsulated in the R language program is used to obtain the optimal threshold point from the ROC curve analysis based on the Youden index.
[0048] Through the analysis of the lightGBM machine learning model, REN is a protein molecule that contributes most to the lightGBM machine learning model.
[0049] Figure 2 The results showed that the lightGBM machine learning model predicted an AUC value of 0.61. In addition, through ROC analysis, the Youden optimal threshold point of REN was 0.476. The present invention uses this as the prediction threshold point for the future risk of myocardial infarction, that is, when the NPX value of REN is ≥ 0.476, the subject has a high risk of myocardial infarction in the future, and when the NPX value of REN is < 0.476, the subject has a low risk of myocardial infarction in the future.
[0050] In this embodiment, 51294 subjects were divided into high-risk group and low-risk group according to their plasma REN levels at the time of first enrollment. Thereafter, the occurrence of myocardial infarction in the respondents was followed up every 2.5 years for a total of 15 years. According to the occurrence of myocardial infarction at each follow-up, the survival curves of the two risk groups were drawn. The high-risk group refers to a high risk of myocardial infarction in the future, and the low-risk group refers to a low risk of myocardial infarction in the future: peripheral blood plasma was collected, and the protein level of REN was detected according to the Olink kit, and the protein level of REN was converted into NPX value, and then the high-risk group and low-risk group were determined. The determination method was: when the NPX value of REN was ≥0.476, the risk of myocardial infarction in the future of the subject was high, and when the NPX value of REN was <0.476, the risk of myocardial infarction in the future of the subject was low. Then the survival rate of the high-risk group and low-risk group samples was analyzed separately using Kaplan-Meier survival analysis.
[0051] Figure 3 The results showed that the average risk of myocardial infarction in the high-risk group was 1.66 times higher than that in the low-risk group. The probability of long-term myocardial infarction in the high-risk group was significantly higher than that in the low-risk group. The statistical test was significant, indicating that the use of REN was a significant risk indicator for myocardial infarction in the subjects within the next 15 years.
[0052] Example 2
[0053] The expression level of REN protein is used to assess the risk of myocardial infarction in the future
[0054] 1.1 Study subjects and clinical data
[0055] This example included a total of 100 subjects with an average age of 60.5 years who were admitted to our hospital between 2005 and 2012, including 48 males and 52 females. All subjects had not experienced myocardial infarction. The occurrence of myocardial infarction within 10 years was continuously followed up.
[0056] 1.1.1 The diagnosis of myocardial infarction is made jointly by two cardiologists with more than 10 years of clinical experience. The main basis includes the following aspects:
[0057] (1) Biochemical indicators of myocardial injury:
[0058] Patterns of increases and decreases in plasma cardiac markers such as troponin, especially troponin levels exceeding the upper limit of the 99th percentile reference value.
[0059] (2) Clinical manifestations:
[0060] Typical angina or precordial pain that lasts for more than 20 minutes and is not completely relieved by rest or nitroglycerin.
[0061] It may be accompanied by other symptoms such as sweating, nausea, vomiting, and difficulty breathing.
[0062] (3) Electrocardiogram (ECG) changes:
[0063] T segment elevation or depression, new left bundle branch block (LBBB), pathological Q wave, etc.
[0064] Dynamic T wave changes.
[0065] (4) Imaging evidence:
[0066] Echocardiography, cardiac magnetic resonance imaging (MRI), etc. showed new local ventricular wall motion abnormalities.
[0067] 1.1.2 Exclusion criteria include:
[0068] (1) Non-cardiac chest pain:
[0069] Chest pain caused by other reasons (such as gastroesophageal reflux disease, pleurisy, etc.).
[0070] (2) Other major diseases:
[0071] Recent history of severe heart disease (such as heart failure, heart surgery).
[0072] Active infection or other systemic disease (such as cancer, severe liver or kidney dysfunction).
[0073] (3) Medication or therapeutic intervention:
[0074] Patients who have recently received anticoagulant therapy or have a bleeding tendency.
[0075] Patients who have received thrombolysis or coronary intervention.
[0076] 1.2 Sample collection
[0077] Peripheral blood of the above 100 subjects was collected using EDTA tubes and centrifuged at 2500 g for 10 min at 4°C. After centrifugation, the supernatant was collected to obtain plasma.
[0078] 1.3 Detection of REN concentration in plasma using Olink kit
[0079] By using Olink kit (purchased from Thermo Fisher Scientific, product name Explore3072 / 384) detected the REN concentration in the subjects' plasma, and the measured REN concentration was calibrated and converted into the NPX value using the NPX method to obtain the NPX value of REN to ensure the accuracy of the results.
[0080] 1.4 Criteria
[0081] When the NPX value of REN is ≥0.476, the subject has a high risk of suffering myocardial infarction in the future; when the NPX value of REN is <0.476, the subject has a low risk of suffering myocardial infarction in the future.
[0082] 1.5 Data analysis and statistics using REN’s NPX value
[0083] The subjects are divided into a high-risk group and a low-risk group according to their risk of future myocardial infarction. The high-risk group refers to subjects with a high risk of future myocardial infarction, and the low-risk group refers to subjects with a low risk of future myocardial infarction.
[0084] According to the above criteria, 43 of the 100 subjects were classified into the high-risk group and 57 into the low-risk group.
[0085] Follow-up and data collection: All subjects were followed up for 10 years, and the occurrence of myocardial infarction during this period was recorded.
[0086] Data results:
[0087] During the follow-up period, 5 patients in the high-risk group developed myocardial infarction, while no patients in the low-risk group developed myocardial infarction.
[0088] Statistical analysis: The chi-square test was used to calculate the P value and analyze the association between REN level and myocardial infarction risk. The p value was 0.013, indicating that REN has a significant risk prediction effect. Since the number of patients in the low-risk group was 0, the OR value was infinite.
[0089] The results showed that the above judgment criteria can be used to well distinguish the risk of myocardial infarction in the next 10 years by detecting REN concentration. The risk of myocardial infarction in the next 10 years in the high-risk group was significantly higher than that in the low-risk group. This finding shows the value of REN as a potential biomarker for myocardial infarction risk assessment and supports its application in cardiovascular disease prevention strategies.
[0090] 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 principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.
Claims
1. Use of a reagent for detecting REN protein in the preparation of a product for assessing a subject's future risk of myocardial infarction.
2. The use according to claim 1, characterized in that: The reagent for detecting REN protein includes a reagent for detecting the expression amount of REN protein.
3. The use according to claim 2, characterized in that: The reagents include antibodies to the REN protein.
4. The use according to claim 3, characterized in that: The reagents include an enzyme-linked immunosorbent assay kit.
5. The use according to claim 1, characterized in that: The future time is more than 10 years.
6. The use according to claim 1, characterized in that: The sample from the subject includes peripheral blood.
7. The use according to claim 1, characterized in that: The product comprises a reagent or a kit.
8. The use according to claim 1, characterized in that: The evaluation standard is that when the NPX value of REN is ≥ 0.476, the subject has a high risk of suffering from myocardial infarction in the future; when the NPX value of REN is < 0.476, the subject has a low risk of suffering from myocardial infarction in the future.
9. A system for assessing a subject's risk of future myocardial infarction, characterized in that: It comprises a detection module and a judgment module, wherein the detection module is used to obtain the REN protein concentration in plasma and convert the REN protein concentration into the REN protein NPX value, and the judgment module comprises a readable carrier storing the judgment standard.
10. The system according to claim 9, characterized in that The judgment standard is that when the NPX value of REN is ≥ 0.476, the risk of the subject suffering from myocardial infarction in the future is high, and when the NPX value of REN is < 0.476, the risk of the subject suffering from myocardial infarction in the future is low.
