System and kit for diagnosing antiphospholipid syndrome

Abstract

According to the system and the kit for diagnosing the anti-phospholipid syndrome, provided by the invention, whether a patient suffers from the anti-phospholipid syndrome or not can be accurately diagnosed according to the expression quantity of the biomarker, so that a treatment strategy is more accurately and individually selected; a novel, efficient and non-invasive auxiliary tool is provided for clinically formulating an individualized diagnosis and treatment scheme of the anti-phospholipid syndrome.

Description

Technical Field

[0001] The present disclosure relates to the field of biotechnology, and in particular, to a system and a kit for diagnosing antiphospholipid syndrome. Background Art

[0002] Antiphospholipid syndrome (APS) is a group of autoimmune diseases with recurrent arteriovenous thrombosis and / or pregnancy complications as the main clinical manifestations. The clinical manifestations of APS are complex and lack specificity, so it is difficult to diagnose in the early stage. Antiphospholipid antibodies (aPLs) are an important basis for the clinical diagnosis and treatment of APS, and the quality of the detection method is the key to obtaining reliable autoantibody detection results. The antiphospholipid antibodies detected clinically in the classification and diagnostic criteria of APS include lupus anticoagulants, anticardiolipin antibodies, and anti-β2GP1 antibodies. Since other infections, tumor patients and even healthy people may also have antiphospholipid antibodies, this indicator has a certain proportion of interference with diagnosis and is prone to misdiagnosis. In addition, about 40% to 50% of patients with typical clinical manifestations are negative for the three antiphospholipid antibodies detected routinely, often referred to as "serologically negative" APS, and these patients are easily missed. Current studies have found that detecting antiphospholipid antibodies other than lupus anticoagulant (LA), antiphospholipid antibodies (aCL), and anti-β2GP1 antibodies can help diagnose "serologically negative" APS, and the more types of antiphospholipid antibodies detected, the greater the possibility of diagnosing APS.

[0003] At present, there are literatures that evaluate the diagnostic value of single antiphospholipid antibodies, but do not give the value of combined detection. Therefore, it is urgent to provide a model based on the combined detection of antiphospholipid antibodies to improve the diagnosis level of APS patients. Summary of the invention

[0004] The purpose of the present disclosure is to provide a system for the combined diagnosis of antiphospholipid syndrome based on antiphospholipid antibodies, which effectively improves the diagnostic level of patients with antiphospholipid syndrome, thereby more accurately and individually selecting treatment strategies.

[0005] In one aspect, the present disclosure provides a system for diagnosing antiphospholipid syndrome, the system comprising: a computing device, an input device for inputting quantitative detection results of biomarkers, and an output device for outputting antiphospholipid syndrome diagnosis results; The biomarkers include LA, anti-aCL antibody IgG, anti-aCL antibody IgM, anti-aCL antibody IgA, anti-β2GP1 antibody IgG, anti-β2GP1 antibody IgM, anti-β2GP1 antibody IgA, anti-β2GP1 D1 antibody IgG, anti-PC antibody IgG, anti-PC antibody IgM, anti-PS antibody IgG, anti-PS antibody IgM, anti-PT antibody IgG, anti-PT antibody IgM, anti-PT antibody IgA, anti-PE antibody IgG, anti-PE antibody IgM, anti-PI antibody IgG, anti-PI antibody IgM, anti-thrombin antibody IgG, anti-thrombin antibody IgA, anti-thrombin antibody IgM, anti-Annexin V antibody IgG, anti-Annexin V antibody IgM, anti-laminin antibody IgG, anti-Sphingomeylin antibody IgG, anti-Sphingomeylin antibody IgM, anti-PT / PS antibody IgG and a combination of anti-PT / PS antibody IgM; The biomarker expression level refers to the absolute expression level of each biomarker; The computing device includes a memory and a processor, wherein the memory stores a computer program, and the processor is configured to execute the computer program stored in the memory to implement a modeling algorithm based on a random forest model composed of 100 decision trees and an algorithm of a discriminant function as shown in formula (1); F ( x )=majority({ f 1( x ), f 2( x ),…,f 100 ( x )}) Formula (1); In formula (1), F(x) The final judgment result is 1, which means the subject suffers from antiphospholipid syndrome, and 0, which means the subject does not suffer from antiphospholipid syndrome. x represents the absolute expression of each biomarker; f 1( x ), f 2( x ),…,f 100 ( x ) is a kernel function obtained by training a random forest model composed of 100 decision trees; the kernel function f 1( x ), f 2( x ),…,f 100 (x ) determining whether the subject suffers from antiphospholipid syndrome based on the absolute expression level of each biomarker, wherein a return value of 1 indicates that the subject suffers from antiphospholipid syndrome, and a return value of 0 indicates that the subject does not suffer from antiphospholipid syndrome; The majority function is used to describe the kernel function f 1( x ), f 2( x ),…,f 100 ( x ) is used to perform majority voting on the judgment results, count the frequency of occurrence of each judgment result, and take the category with the highest frequency of occurrence as the final judgment result.

[0006] Optionally, the biomarker is derived from a serum sample from a patient with antiphospholipid syndrome.

[0007] The system also includes a detection device for detecting the expression amount of the biomarker; The biomarker detection device includes antigen reagents for respectively detecting the expression amount of each biomarker.

[0008] Optionally, each decision tree is based on the Gini impurity as a segmentation criterion, and the features and segmentation points that minimize the sum of the subset Gini impurities are selected for training, thereby obtaining kernel functions f1(x), f2(x),…,f for determining whether the subject suffers from antiphospholipid syndrome based on the absolute expression of each biomarker. 100 (x).

[0009] On the other hand, the present disclosure provides the use of a reagent for detecting the expression amount of a biomarker in the preparation of a kit for diagnosing antiphospholipid syndrome, wherein the biomarker includes LA, anti-aCL antibody IgG, anti-aCL antibody IgM, anti-aCL antibody IgA, anti-β2GP1 antibody IgG, anti-β2GP1 antibody IgM, anti-β2GP1 antibody IgA, anti-β2GP1 D1 antibody IgG, anti-PC antibody IgG, anti-PC antibody IgM, anti-PS antibody IgG, anti-PS antibody IgM, anti-PT antibody IgG, anti-PT antibody IgM, anti-PT antibody IgA, anti-PE antibody IgG, anti-PE antibody IgM, anti-PI antibody IgG, anti-PI antibody IgM, anti-thrombin antibody IgG, anti-thrombin antibody IgA, anti-thrombin antibody IgM, anti-Annexin V antibody IgG, anti-Annexin V antibody IgM, anti-laminin antibody IgG, anti-Sphingomeylin antibody IgG, anti-Sphingomeylin antibody IgM, anti-PT / PS antibody IgG and anti-PT / PS Combination of antibodies IgM.

[0010] On the other hand, the present disclosure provides a kit for diagnosing antiphospholipid syndrome, the kit comprising reagents for quantitatively detecting the expression amount of biomarkers, wherein the biomarkers include LA, anti-aCL antibody IgG, anti-aCL antibody IgM, anti-aCL antibody IgA, anti-β2GP1 antibody IgG, anti-β2GP1 antibody IgM, anti-β2GP1 antibody IgA, anti-β2GP1 D1 antibody IgG, anti-PC antibody IgG, anti-PC antibody IgM, anti-PS antibody IgG, anti-PS antibody IgM, anti-PT antibody IgG, anti-PT antibody IgM, anti-PT antibody IgA, anti-PE antibody IgG, anti-PE antibody IgM, anti-PI antibody IgG, anti-PI antibody IgM, anti-thrombin antibody IgG, anti-thrombin antibody IgA, anti-thrombin antibody IgM, anti-Annexin V antibody IgG, anti-Annexin V antibody IgM, anti-laminin antibody IgG, anti-Sphingomeylin antibody IgG, anti-Sphingomeylin antibody IgM, anti-PT / PS antibody IgG, and anti-PT / PS Combination of antibodies IgM.

[0011] Optionally, the process of quantitatively detecting the expression of biomarkers includes: S1. Obtain serum samples from patients with antiphospholipid syndrome; S2. Determine the expression level of the biomarker in the serum sample.

[0012] On the other hand, the present disclosure provides use of biomarkers in preparing a kit for diagnosing antiphospholipid syndrome, wherein the biomarkers include LA, anti-aCL antibody IgG, anti-aCL antibody IgM, anti-aCL antibody IgA, anti-β2GP1 antibody IgG, anti-β2GP1 antibody IgM, anti-β2GP1 antibody IgA, anti-β2GP1 D1 antibody IgG, anti-PC antibody IgG, anti-PC antibody IgM, anti-PS antibody IgG, anti-PS antibody IgM, anti-PT antibody IgG, anti-PT antibody IgM, anti-PT antibody IgA, anti-PE antibody IgG, anti-PE antibody IgM, anti-PI antibody IgG, anti-PI antibody IgM, anti-thrombin antibody IgG, anti-thrombin antibody IgA, anti-thrombin antibody IgM, anti-Annexin V antibody IgG, anti-Annexin V antibody IgM, anti-laminin antibody IgG, anti-Sphingomeylin antibody IgG, anti-Sphingomeylin antibody IgM, anti-PT / PS antibody IgG, and a combination of anti-PT / PS antibody IgM.

[0013] Through the above technical scheme, the present disclosure provides a system and a kit for diagnosing antiphospholipid syndrome, which can accurately diagnose whether a patient has antiphospholipid syndrome based on the expression level of biomarkers, thereby more accurately and individually selecting treatment strategies, and providing a new, efficient and non-invasive auxiliary tool for the clinical formulation of individualized diagnosis and treatment plans for antiphospholipid syndrome.

[0014] Other features and advantages of the present disclosure will be described in detail in the following detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the following specific embodiments, they are used to explain the present invention but do not constitute a limitation of the present invention. In the accompanying drawings: Figure 1 ROC curves for different models to judge diseases.

[0016] Figure 2 Precision and recall curves of the random forest model in the disease classification task.

[0017] Figure 3 Decision diagram for the random forest model.

[0018] Figure 4 The importance of different antiphospholipid antibodies in the random forest model.

[0019] Figure 5 is one decision tree among 100 decision tree models.

[0020] Figure 6 ROC curves of random forest model and single antibody.

[0021] Figure 7 Characteristic spectrum of antiphospholipid antibodies in different clinical subtypes. DETAILED DESCRIPTION

[0022] The specific embodiments of the present disclosure are described in detail below. It should be understood that the specific embodiments described herein are only used to illustrate and explain the present disclosure, and are not used to limit the present disclosure.

[0023] The present disclosure provides a system for diagnosing antiphospholipid syndrome, the system comprising: a computing device, an input device for inputting quantitative detection results of biomarkers, and an output device for outputting antiphospholipid syndrome diagnosis results; The biomarkers include LA, anti-aCL antibody IgG, anti-aCL antibody IgM, anti-aCL antibody IgA, anti-β2GP1 antibody IgG, anti-β2GP1 antibody IgM, anti-β2GP1 antibody IgA, anti-β2GP1 D1 antibody IgG, anti-PC antibody IgG, anti-PC antibody IgM, anti-PS antibody IgG, anti-PS antibody IgM, anti-PT antibody IgG, anti-PT antibody IgM, anti-PT antibody IgA, anti-PE antibody IgG, anti-PE antibody IgM, anti-PI antibody IgG, anti-PI antibody IgM, anti-thrombin antibody IgG, anti-thrombin antibody IgA, anti-thrombin antibody IgM, anti-Annexin V antibody IgG, anti-Annexin V antibody IgM, anti-laminin antibody IgG, anti-Sphingomeylin antibody IgG, anti-Sphingomeylin antibody IgM, anti-PT / PS antibody IgG and a combination of anti-PT / PS antibody IgM; The biomarker expression level refers to the absolute expression level of each biomarker; The computing device includes a memory and a processor, wherein the memory stores a computer program, and the processor is configured to execute the computer program stored in the memory to implement a modeling algorithm based on a random forest model composed of 100 decision trees and an algorithm of a discriminant function as shown in formula (1); F ( x )=majority({ f 1( x ), f 2( x ),…,f 100 ( x)}) Formula (1); In formula (1), F(x) is the final judgment result. The return value of 1 indicates that the subject suffers from antiphospholipid syndrome. F(x) A return value of 0 indicates that the subject does not have antiphospholipid syndrome. x represents the absolute expression of each biomarker; f 1( x ), f 2( x ),…,f 100 ( x ) is a kernel function obtained by training a random forest model composed of 100 decision trees; the kernel function f 1( x ), f 2( x ),…,f 100 ( x ) determining whether the subject suffers from antiphospholipid syndrome based on the absolute expression level of each biomarker, wherein a return value of 1 indicates that the subject suffers from antiphospholipid syndrome, and a return value of 0 indicates that the subject does not suffer from antiphospholipid syndrome; The majority function is used to describe the kernel function f 1( x ), f 2( x ),…,f 100 ( x ) is used to perform majority voting on the judgment results, count the frequency of occurrence of each judgment result, and take the category with the highest frequency of occurrence as the final judgment result.

[0024] Optionally, the biomarker is derived from a serum sample from a patient with antiphospholipid syndrome.

[0025] The system also includes a detection device for detecting the expression amount of the biomarker; The biomarker detection device includes antigen reagents for respectively detecting the expression amount of each biomarker.

[0026] Optionally, each decision tree is based on the Gini impurity as a segmentation criterion, and the features and segmentation points that minimize the sum of the subset Gini impurities are selected for training, thereby obtaining kernel functions f1(x), f2(x),…,f for determining whether the subject suffers from antiphospholipid syndrome based on the absolute expression of each biomarker. 100 (x).

[0027] On the other hand, the present disclosure provides the use of a reagent for detecting the expression amount of a biomarker in the preparation of a kit for diagnosing antiphospholipid syndrome, wherein the biomarker includes LA, anti-aCL antibody IgG, anti-aCL antibody IgM, anti-aCL antibody IgA, anti-β2GP1 antibody IgG, anti-β2GP1 antibody IgM, anti-β2GP1 antibody IgA, anti-β2GP1 D1 antibody IgG, anti-PC antibody IgG, anti-PC antibody IgM, anti-PS antibody IgG, anti-PS antibody IgM, anti-PT antibody IgG, anti-PT antibody IgM, anti-PT antibody IgA, anti-PE antibody IgG, anti-PE antibody IgM, anti-PI antibody IgG, anti-PI antibody IgM, anti-thrombin antibody IgG, anti-thrombin antibody IgA, anti-thrombin antibody IgM, anti-Annexin V antibody IgG, anti-Annexin V antibody IgM, anti-laminin antibody IgG, anti-Sphingomeylin antibody IgG, anti-Sphingomeylin antibody IgM, anti-PT / PS antibody IgG and anti-PT / PS Combination of antibodies IgM.

[0028] On the other hand, the present disclosure provides a kit for diagnosing antiphospholipid syndrome, the kit comprising reagents for quantitatively detecting the expression amount of biomarkers, wherein the biomarkers include LA, anti-aCL antibody IgG, anti-aCL antibody IgM, anti-aCL antibody IgA, anti-β2GP1 antibody IgG, anti-β2GP1 antibody IgM, anti-β2GP1 antibody IgA, anti-β2GP1 D1 antibody IgG, anti-PC antibody IgG, anti-PC antibody IgM, anti-PS antibody IgG, anti-PS antibody IgM, anti-PT antibody IgG, anti-PT antibody IgM, anti-PT antibody IgA, anti-PE antibody IgG, anti-PE antibody IgM, anti-PI antibody IgG, anti-PI antibody IgM, anti-thrombin antibody IgG, anti-thrombin antibody IgA, anti-thrombin antibody IgM, anti-Annexin V antibody IgG, anti-Annexin V antibody IgM, anti-laminin antibody IgG, anti-Sphingomeylin antibody IgG, anti-Sphingomeylin antibody IgM, anti-PT / PS antibody IgG, and anti-PT / PS Combination of antibodies IgM.

[0029] Optionally, the process of quantitatively detecting the expression of biomarkers includes: S1. Obtain serum samples from patients with antiphospholipid syndrome; S2. Determine the expression level of the biomarker in the serum sample.

[0030] On the other hand, the present disclosure provides use of biomarkers in preparing a kit for diagnosing antiphospholipid syndrome, wherein the biomarkers include LA, anti-aCL antibody IgG, anti-aCL antibody IgM, anti-aCL antibody IgA, anti-β2GP1 antibody IgG, anti-β2GP1 antibody IgM, anti-β2GP1 antibody IgA, anti-β2GP1 D1 antibody IgG, anti-PC antibody IgG, anti-PC antibody IgM, anti-PS antibody IgG, anti-PS antibody IgM, anti-PT antibody IgG, anti-PT antibody IgM, anti-PT antibody IgA, anti-PE antibody IgG, anti-PE antibody IgM, anti-PI antibody IgG, anti-PI antibody IgM, anti-thrombin antibody IgG, anti-thrombin antibody IgA, anti-thrombin antibody IgM, anti-Annexin V antibody IgG, anti-Annexin V antibody IgM, anti-laminin antibody IgG, anti-Sphingomeylin antibody IgG, anti-Sphingomeylin antibody IgM, anti-PT / PS antibody IgG, and a combination of anti-PT / PS antibody IgM.

[0031] The present invention is further described in detail below by way of examples.

[0032] Example 1 This example is used to illustrate the screening of anti-phospholipid antibodies, model screening and system construction.

[0033] Sample screening: Serum samples of 86 antiphospholipid syndrome patients and 143 healthy control patients were randomly selected (all from Peking Union Medical College Hospital); 70% of them were used as training sets and 30% as test sets. The training set included 60 antiphospholipid syndrome patients and 100 control samples, and the test set included 26 antiphospholipid syndrome patients and 43 control samples.

[0034] Sample testing: (1) LA was tested using the HemosIL®DRVVT screening and confirmation kit. A DRVVT standardized ratio ≥1.2 was considered positive and set to 1, whereas a DRVVT standardized ratio below 1.2 was considered negative and set to 0; (2) Anti-aCL antibody IgG, anti-aCL antibody IgM, anti-aCL antibody IgA, anti-β2GP1 antibody IgG, anti-β2GP1 antibody IgM, anti-β2GP1 antibody IgA and anti-β2GP1 D1 antibody IgG were detected using the QUANTA Flash chemiluminescence kit. If the antibody concentration level is ≥ 20 CU, it is judged as positive and set to 1, otherwise it is judged as negative and set to 0; (3) Anti-PC antibody IgG, anti-PC antibody IgM, anti-PS antibody IgG, anti-PS antibody IgM, anti-PT antibody IgG, anti-PT antibody IgM, anti-PT antibody IgA, anti-PE antibody IgG, anti-PE antibody IgM, anti-PI antibody IgG, anti-PI antibody IgM, anti-thrombin antibody IgG, anti-thrombin antibody IgA, anti-thrombin antibody IgM, anti-Annexin V antibody IgG, anti-Annexin V antibody IgM, anti-laminin antibody IgG, anti-Sphingomeylin antibody IgG, anti-Sphingomeylin antibody IgA were detected using the enzyme-linked immunosorbent assay kit of AESKU. If the antibody concentration level is >18U / mL, it is judged as positive and set to 1; otherwise, it is judged as negative and set to 0; (4) Anti-PT / PS antibody IgG and anti-PT / PS antibody IgM were detected using the QUANTA Lite enzyme-linked immunosorbent assay kit. If the antibody concentration level was >30 units, it was considered positive and set to 1; otherwise, it was considered negative and set to 0.

[0035] During the above testing process, strictly follow the operating instructions of the test kit to ensure the accuracy and repeatability of the test results; Based on the above sample measurement results, each modeling algorithm model is evaluated: In order to screen out the most suitable model, this embodiment further compares the performance of five different models, namely, logistic regression model, random forest model, support vector machine (SVM), gradient boosting machine (GBM), and multilayer perceptron (MLP) neural network. The above training set and test set are used to train and evaluate these models respectively. The final accuracy comparison results of different models are shown in Table 1. The ROC curves of different models for judging disease classification are shown in Table 1. Figure 1 shown.

[0036] Table 1 Accuracy of different models Model Accuracy Logistic Regression 89.86% Random Forest 94.20% Support Vector Machine (SVM) 69.57% Gradient Boosting Machine (GBM) 94.20% Multilayer Perceptron (MLP) Neural Network 59.42% Among them, the accuracy of the random model is 94.20%. For category 1 (APS), its precision is 98%, recall is 93%, and F1 score is 95%. For category 0 (control), its precision is 89%, recall is 96%, and F1 score is 93%. The precision and recall curve of the random forest model in the disease classification task shows the accuracy of the model in predicting the positive class at different recall rates. The closer the area under the curve (AUC) value is to 1, the better the performance of the model. Figure 2 As can be seen from the figure, the precision and recall curves of the random forest model are above the random guessing line, indicating that the model performs well in the disease classification task and has good performance in distinguishing APS from controls. The decision diagram of the random forest model is shown in Figure 3 The importance of different antiphospholipid antibodies was evaluated using a random forest model. Figure 4 As shown in the figure, the greater the absolute value of the importance score, the greater the impact of antiphospholipid antibodies in the model. The algorithm flow chart of the random forest model is as follows: Figure 5 shown.

[0037] Table 2 Importance scores of different antiphospholipid antibodies in the random forest model

[0038] Based on the accuracy of the above random model, the present disclosure selects a random forest model for testing, randomly divides the antibody expression data in the above samples into two data sets, respectively as a training set and a validation set, performs preprocessing on the data, removes missing values ​​in the data set to ensure the integrity and accuracy of the data, identifies and removes outliers to avoid interference with subsequent analysis, and removes duplicate data to ensure the uniqueness of the data. The features are standardized using the Z-score standardization method to eliminate the impact of different features due to dimensional differences. In terms of parameter settings, the number of trees (n_estimators) is set to 100, the maximum depth (max_depth) is set to None, the minimum number of split samples (min_samples_split) is set to 2, the minimum number of leaf node samples (min_samples_leaf) is set to 1, and the random state (random_state) is set to 42; The calculation coefficients and thresholds in the training system are trained using the training set, and the system is evaluated using the validation set. The discriminant function of the system is shown in formula (1): The computing device includes a memory and a processor, wherein the memory stores a computer program, and the processor is configured to execute the computer program stored in the memory to implement a modeling algorithm based on a random forest model composed of 100 decision trees and an algorithm of a discriminant function as shown in formula (1); F ( x )=majority({ f 1( x ), f 2( x ),…,f 100 ( x )}) Formula (1); in, F(x) A return value of 1 indicates that the subject has antiphospholipid syndrome. F(x) A return value of 0 indicates that the subject does not have antiphospholipid syndrome. x represents the absolute expression of each biomarker; f 1( x ), f 2( x ),…,f 100 ( x ) is a kernel function obtained by training a random forest model composed of 100 decision trees; the kernel function f 1( x ), f 2( x ),…,f 100 ( x ) determines whether the subject suffers from antiphospholipid syndrome based on the absolute expression level of each biomarker, and the return value of 1 indicates that the subject suffers from antiphospholipid syndrome, and the return value of 0 indicates that the subject does not suffer from antiphospholipid syndrome; the majority function is applied to the kernel function f 1( x ), f 2( x ),…,f 100 ( x ) is used to perform majority voting on the judgment results, count the frequency of occurrence of each judgment result, and take the category with the highest frequency of occurrence as the final judgment result.

[0039] Each decision tree is based on the Gini impurity as the segmentation criterion, and the features and segmentation points that minimize the sum of the subset Gini impurities are selected for training, thereby obtaining the kernel functions f1(x), f2(x),…,f for determining whether the subject has antiphospholipid syndrome based on the absolute expression of each biomarker. 100 (x), the calculation formula of Gini impurity is: Formula (2); in, D is a dataset, p k is the first k The relative frequency of the class.

[0040] Example 2 This example further verifies the accuracy of the system in Example 1 for diagnosing APS and measures the ROC curve.

[0041] The performance of the system constructed by the system of the present invention in the disease classification task was evaluated by the ROC (Receiver Operating Characteristic) curve, showing that its area under the curve (AUC) was 0.98, indicating that its diagnostic performance was far superior to the single indicators of routine clinical tests (anti-PT / PS antibody IgG, anti-PT / PS antibody IgM, LA, anti-β2GP1 D1 antibody IgG, anti-aCL antibody IgA, anti-β2GP1 antibody IgA, anti-aCL antibody IgG, anti-β2GP1 antibody IgG, anti-aCL antibody IgM, anti-β2GP1 antibody IgM). The use of the system of the present invention for diagnosing APS significantly improved the diagnostic efficacy of APS. Figure 6 The ROC curve of each antiphospholipid antibody in is represented by a dotted line of a different color, and the ROC curve of the system of the present disclosure is represented by a solid line.

[0042] Compared with anti-PT / PS antibody IgG, anti-PT / PS antibody IgM, LA, anti-β2GP1 D1 antibody IgG, anti-aCL antibody IgA, anti-β2GP1 antibody IgA, anti-aCL antibody IgG, anti-β2GP1 antibody IgG, anti-aCL antibody IgM, and anti-β2GP1 antibody IgM, the system disclosed in the present invention has the highest sensitivity and better specificity for diagnosing antiphospholipid syndrome (as shown in Table 3).

[0043] Table 3. Sensitivity and specificity of random forest model and single antiphospholipid antibody in diagnosing APS

[0044] The combined detection of multiple antiphospholipid antibodies is helpful to distinguish people with different clinical manifestations of arterial thrombosis, venous thrombosis and morbid pregnancy. Cluster analysis can divide the antiphospholipid syndrome patients into three subgroups with different clinical manifestations and different antiphospholipid antibody spectra (as shown in Table 4 and Figure 7 as shown).

[0045] Table 4. Clinical manifestations and antiphospholipid antibody spectrum of different clinical subtypes

[0046] Example 3 The random forest model was substituted into the system to determine whether 30% of the patients (a total of 69 cases, including 26 patients with antiphospholipid syndrome and 43 control samples) used for testing had antiphospholipid syndrome, and then compared with the clinical diagnosis results of the patients. The comparison results are shown in Table 1 below, where 1 indicates that the patient has antiphospholipid syndrome and 0 indicates that the patient does not have antiphospholipid syndrome: Table 5 Comparison of the patient's clinical diagnosis and system prediction diagnosis results

[0047] Taking clinical diagnosis as the gold standard, the specificity of the system's predicted diagnosis was 93.02%, the sensitivity was 96.15%, and the accuracy was 94.20%, indicating that the disclosed system performed well in disease judgment and could effectively identify patients with antiphospholipid syndrome.

[0048] Through the above embodiments, the present invention assigns weight values ​​to different antiphospholipid antibodies through a modeling algorithm, and constructs a system for diagnosing antiphospholipid syndrome based on the weight values ​​of different antiphospholipid antibodies, so as to accurately diagnose whether the patient suffers from antiphospholipid syndrome, thereby more accurately performing individualized treatment, with high specificity and sensitivity, and providing a new, efficient and non-invasive auxiliary tool for the clinical formulation of individualized diagnosis and treatment plans for antiphospholipid syndrome.

[0049] The preferred embodiments of the present disclosure are described in detail above; however, the present disclosure is not limited to the specific details in the above embodiments. Within the technical concept of the present disclosure, a variety of simple modifications can be made to the technical solution of the present disclosure, and these simple modifications all fall within the protection scope of the present disclosure. It should also be noted that the various specific technical features described in the above specific embodiments can be combined in any suitable manner without contradiction. In order to avoid unnecessary repetition, the present disclosure will not further describe various possible combinations.

[0050] In addition, various embodiments of the present disclosure may be arbitrarily combined, and as long as they do not violate the concept of the present disclosure, they should also be regarded as the contents disclosed by the present disclosure.

Claims

1. A system for diagnosing antiphospholipid syndrome, characterized in that: The system comprises: a computing device, an input device for inputting the expression amount of a biomarker, and an output device for outputting a diagnosis result of antiphospholipid syndrome; Wherein, the biomarkers include LA, anti-aCL antibody IgG, anti-aCL antibody IgM, anti-aCL antibody IgA, anti-β2GP1 antibody IgG, anti-β2GP1 antibody IgM, anti-β2GP1 antibody IgA, anti-β2GP1 D1 antibody IgG, anti-PC antibody IgG, anti-PC antibody IgM, anti-PS antibody IgG, anti-PS antibody IgM, anti-PT antibody IgG, anti-PT antibody IgM, anti-PT antibody IgA, anti-PE antibody IgG, anti-PE antibody IgM, anti-PI antibody IgG, anti-PI antibody IgM, anti-thrombin antibody IgG, anti-thrombin antibody IgA, anti-thrombin antibody IgM, anti-Annexin V antibody IgG, anti-Annexin V antibody IgM, anti-laminin antibody IgG, anti-Sphingomeylin antibody IgG, anti-Sphingomeylin antibody IgM, anti-PT / PS antibody IgG and a combination of anti-PT / PS antibody IgM; The biomarker expression level refers to the absolute expression level of each biomarker; The computing device includes a memory and a processor, wherein the memory stores a computer program, and the processor is configured to execute the computer program stored in the memory to implement a modeling algorithm based on a random forest model composed of 100 decision trees and an algorithm of a discriminant function as shown in formula (1); F ( x )=majority({ f 1( x ), f 2( x ),…,f 100 ( x )}) Formula (1); In formula (1), F(x) The final judgment result is 1, which means the subject suffers from antiphospholipid syndrome, and 0, which means the subject does not suffer from antiphospholipid syndrome. x represents the absolute expression of each biomarker; f 1( x ), f 2( x ),…,f 100 ( x ) is a kernel function obtained by training a random forest model composed of 100 decision trees; the kernel function f 1( x ), f 2( x ),…,f 100 ( x ) determining whether the subject suffers from antiphospholipid syndrome based on the absolute expression level of each biomarker, wherein a return value of 1 indicates that the subject suffers from antiphospholipid syndrome, and a return value of 0 indicates that the subject does not suffer from antiphospholipid syndrome; The majority function is used to describe the kernel function f 1( x ), f 2( x ),…,f 100 ( x ) is used to perform majority voting on the judgment results, count the frequency of occurrence of each judgment result, and take the category with the highest frequency of occurrence as the final judgment result.

2. The system according to claim 1, wherein: The biomarkers are derived from serum samples of patients with antiphospholipid syndrome.

3. The system according to claim 1, wherein: The system also includes a detection device for detecting the expression amount of the biomarker; The biomarker detection device includes antigen reagents for respectively detecting the expression amount of each biomarker.

4. The system according to claim 1, wherein: Each decision tree is based on the Gini impurity as the segmentation criterion, and the features and segmentation points that minimize the sum of the subset Gini impurities are selected for training, thereby obtaining the kernel functions f1(x), f2(x),…,f for determining whether the subject has antiphospholipid syndrome based on the absolute expression of each biomarker. 100 (x).

5. Use of a reagent for detecting the expression of a biomarker in the preparation of a kit for diagnosing antiphospholipid syndrome, characterized in that: The biomarkers include LA, anti-aCL antibody IgG, anti-aCL antibody IgM, anti-aCL antibody IgA, anti-β2GP1 antibody IgG, anti-β2GP1 antibody IgM, anti-β2GP1 antibody IgA, anti-β2GP1 D1 antibody IgG, anti-PC antibody IgG, anti-PC antibody IgM, anti-PS antibody IgG, anti-PS antibody IgM, anti-PT antibody IgG, anti-PT antibody IgM, anti-PT antibody IgA, anti-PE antibody IgG, anti-PE antibody IgM, anti-PI antibody IgG, anti-PI antibody IgM, anti-thrombin antibody IgG, anti-thrombin antibody IgA, anti-thrombin antibody IgM, anti-Annexin V antibody IgG, anti-Annexin V antibody IgM, anti-laminin antibody IgG, anti-Sphingomeylin antibody IgG, anti-Sphingomeylin antibody IgM, anti-PT / PS antibody IgG and a combination of anti-PT / PS antibody IgM.

6. A kit for diagnosing antiphospholipid syndrome, characterized in that: The kit includes reagents for quantitatively detecting the expression amount of biomarkers, and the biomarkers include LA, anti-aCL antibody IgG, anti-aCL antibody IgM, anti-aCL antibody IgA, anti-β2GP1 antibody IgG, anti-β2GP1 antibody IgM, anti-β2GP1 antibody IgA, anti-β2GP1 D1 antibody IgG, anti-PC antibody IgG, anti-PC antibody IgM, anti-PS antibody IgG, anti-PS antibody IgM, anti-PT antibody IgG, anti-PT antibody IgM, anti-PT antibody IgA, anti-PE antibody IgG, anti-PE antibody IgM, anti-PI antibody IgG, anti-PI antibody IgM, anti-thrombin antibody IgG, anti-thrombin antibody IgA, anti-thrombin antibody IgM, anti-Annexin V antibody IgG, anti-Annexin V antibody IgM, anti-laminin antibody IgG, anti-Sphingomeylin antibody IgG, anti-Sphingomeylin antibody IgM, anti-PT / PS antibody IgG and a combination of anti-PT / PS antibody IgM.

7. The kit according to claim 6, wherein The process of quantitatively detecting the expression of biomarkers includes: S1. Obtain serum samples from patients with antiphospholipid syndrome; S2. Determine the expression level of the biomarker in the serum sample.

8. Use of a biomarker in the preparation of a kit for diagnosing antiphospholipid syndrome, characterized in that: The biomarkers include LA, anti-aCL antibody IgG, anti-aCL antibody IgM, anti-aCL antibody IgA, anti-β2GP1 antibody IgG, anti-β2GP1 antibody IgM, anti-β2GP1 antibody IgA, anti-β2GP1 D1 antibody IgG, anti-PC antibody IgG, anti-PC antibody IgM, anti-PS antibody IgG, anti-PS antibody IgM, anti-PT antibody IgG, anti-PT antibody IgM, anti-PT antibody IgA, anti-PE antibody IgG, anti-PE antibody IgM, anti-PI antibody IgG, anti-PI antibody IgM, anti-thrombin antibody IgG, anti-thrombin antibody IgA, anti-thrombin antibody IgM, anti-Annexin V antibody IgG, anti-Annexin V antibody IgM, anti-laminin antibody IgG, anti-Sphingomeylin antibody IgG, anti-Sphingomeylin antibody IgM, anti-PT / PS antibody IgG and a combination of anti-PT / PS antibody IgM.

Images