A method and system for predicting hyperthyroidism-induced arrhythmia

Abstract

The present invention belongs to the field of intelligent medical technology. The present invention discloses a prediction method and system for hyperthyroid arrhythmia. The present invention discovers for the first time the correlation between gluten exorphin B5 and the diagnosis of hyperthyroid arrhythmia. Based on this, the present invention pioneeringly develops an effective prediction method for hyperthyroid arrhythmia, and provides a related prediction system, computer equipment and computer-readable storage medium, which provides information for early intervention treatment of hyperthyroid arrhythmia and has broad application prospects.

Description

Technical Field

[0001] The present invention belongs to the field of intelligent medical technology, and specifically, relates to a method and system for predicting hyperthyroidism-related arrhythmia. Background Art

[0002] Hyperthyroid heart disease (HHD) is a serious cardiac complication caused by the direct toxic effects or indirect effects of excessive thyroid hormones (THs) on the heart during hyperthyroidism (hyperthyroidism). Hyperthyroid arrhythmias (hyperthyroid arrhythmias) are the most common manifestation of hyperthyroid heart disease, with about 50%-90% of patients with hyperthyroid heart disease experiencing arrhythmias, 85%-100% of patients experiencing tachycardia, and 10%-20% of patients experiencing atrial fibrillation.

[0003] Hyperthyroidism-induced arrhythmias have a long course and poor prognosis, greatly affecting the patient's quality of life and even endangering their lives. The molecular and histological mechanisms of the significant arrhythmia tendency in hyperthyroid heart disease are very complex and not yet fully understood. Studies have shown that excessive thyroid hormones in the blood can cause toxic effects on the myocardium in direct or indirect ways, causing changes in cardiac electrical remodeling in patients with hyperthyroid heart disease, leading to the occurrence of arrhythmias.

[0004] If hyperthyroid arrhythmia can be diagnosed early before irreversible damage occurs, the disease can be evaluated and intervention treatment can be carried out as soon as possible, which can reduce the difficulty of treatment. Therefore, early and accurate diagnosis of hyperthyroid arrhythmia is of great significance. Gluten Exorphin B5 (GE-B5) is an oligopeptide metabolite in the blood. At present, there is no relevant research or report on the application of gluten exorphin B5 in the prediction method of hyperthyroid arrhythmia. Summary of the invention

[0005] In view of this, an object of the present invention is to provide a method, system, electronic device and storage medium for predicting hyperthyroidism-related arrhythmia.

[0006] The present invention aims to solve the technical problem of the lack of an efficient method for predicting hyperthyroid arrhythmia in the prior art. To this end, the present invention provides a method, system, electronic device and storage medium for predicting hyperthyroid arrhythmia. The present invention can predict whether the subject from which the sample to be tested is derived suffers from hyperthyroid arrhythmia by detecting the content of the target metabolite in the sample to be tested, thereby providing information for early intervention treatment of hyperthyroid arrhythmia.

[0007] The above-mentioned object of the present invention is achieved through the following technical solutions:

[0008] A first aspect of the present invention provides a method for predicting hyperthyroidism-induced arrhythmia, the method comprising:

[0009] Obtaining metabolite content data of the sample to be tested;

[0010] extracting content data of a target metabolite from the metabolite content data, wherein the target metabolite is gluten exorphin B5;

[0011] Performing classification prediction based on the content data of the target metabolite to obtain a classification result of whether the sample to be tested is a hyperthyroidism arrhythmia sample;

[0012] The classification result is obtained based on a prediction model, and the method for constructing the prediction model includes: obtaining the content data of the target metabolite of the training set samples and the clinical characteristics corresponding to the samples, the clinical characteristics including patients with hyperthyroid arrhythmia and healthy people, extracting the content data of the target metabolite in the training set and inputting it into a machine learning model to construct a prediction model, and obtaining a constructed prediction model and threshold;

[0013] If the content of the target metabolite is lower than the threshold value, the classification result of the sample to be tested is a hyperthyroidism arrhythmia sample; if the content of the target metabolite is higher than the threshold value, the classification result of the sample to be tested is a non-hyperthyroidism arrhythmia sample.

[0014] Furthermore, the machine learning model is a linear regression model, a logistic regression model, a Lasso regression model, a Ridge regression model, a linear discriminant analysis model, a random forest model, a nearest neighbor model, a naive Bayes model, a decision tree model, a perceptron model, a neural network model, a support vector machine model, an AdaBoost model, a GBDT model, an XGBoost model, a LightGBM model or a CatBoost model.

[0015] Furthermore, the metabolite content data is metabolite content data obtained by detecting the sample to be tested by mass spectrometry, chromatography, liquid chromatography-mass spectrometry, gas chromatography-mass spectrometry, enzyme-linked immunosorbent assay, fluorescence, nuclear magnetic resonance or spectroscopy.

[0016] Furthermore, the sample to be tested is a blood sample, a serum sample, a plasma sample, a tissue sample or a cell sample.

[0017] A second aspect of the present invention provides a prediction system for hyperthyroidism-induced arrhythmia, the system comprising:

[0018] A data acquisition unit, used to acquire metabolite content data of the sample to be tested;

[0019] A data extraction unit, used for extracting content data of a target metabolite from the metabolite content data, wherein the target metabolite is gluten exorphin B5;

[0020] A result prediction unit, used for performing classification prediction based on the content data of the target metabolite to obtain a classification result of whether the sample to be tested is a hyperthyroidism arrhythmia sample;

[0021] The classification result is obtained based on a prediction model, and the method for constructing the prediction model includes: obtaining the content data of the target metabolite of the training set samples and the clinical characteristics corresponding to the samples, the clinical characteristics including patients with hyperthyroid arrhythmia and healthy people, extracting the content data of the target metabolite in the training set and inputting it into a machine learning model to construct a prediction model, and obtaining a constructed prediction model and threshold;

[0022] If the content of the target metabolite is lower than the threshold value, the classification result of the sample to be tested is a hyperthyroidism arrhythmia sample; if the content of the target metabolite is higher than the threshold value, the classification result of the sample to be tested is a non-hyperthyroidism arrhythmia sample.

[0023] Furthermore, the machine learning model is a linear regression model, a logistic regression model, a Lasso regression model, a Ridge regression model, a linear discriminant analysis model, a random forest model, a nearest neighbor model, a naive Bayes model, a decision tree model, a perceptron model, a neural network model, a support vector machine model, an AdaBoost model, a GBDT model, an XGBoost model, a LightGBM model or a CatBoost model.

[0024] The third aspect of the present invention provides a computer device, which includes a memory and a processor, wherein the memory is used to store program instructions; the processor is used to call the program instructions, and when the program instructions are executed, the method for predicting hyperthyroidism-induced arrhythmia described in the first aspect of the present invention is implemented.

[0025] The fourth aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the method for predicting hyperthyroidism-induced arrhythmia described in the first aspect of the present invention is implemented.

[0026] A fifth aspect of the present invention provides the use of a reagent for detecting the content of gluten exorphin B5 in a sample to be tested in the preparation of a diagnostic product for diagnosing hyperthyroidism-related arrhythmia.

[0027] Furthermore, the reagents include reagents for detecting the content of gluten exorphin B5 in a sample by mass spectrometry, chromatography, liquid chromatography-mass spectrometry, gas chromatography-mass spectrometry, enzyme-linked immunosorbent assay, fluorescence, nuclear magnetic resonance or spectroscopy.

[0028] Furthermore, the diagnostic product includes a detection kit, a detection chip or a detection test strip.

[0029] Furthermore, the sample includes a blood sample, a tissue sample, a serum sample, a plasma sample and / or a cell sample.

[0030] In the present invention, the gluten exorphin B5 (GE-B5) is an oligopeptide metabolite in the blood, and its corresponding PubChem CID is 9938501.

[0031] In the present invention, the diagnosis refers to the identification or classification of a molecular or pathological state, disease or condition. For example, by using molecular features (e.g., specific metabolites, specific genes, proteins encoded by specific genes, specific microbial communities), it is possible to differentially diagnose whether a subject is a patient with hyperthyroidism-induced arrhythmia or the risk of the subject suffering from hyperthyroidism-induced arrhythmia.

[0032] In some embodiments, the reagent for detecting the content of gluten exorphin B5 in the sample may also be an antibody that specifically binds to gluten exorphin B5, an affinity protein that specifically binds to gluten exorphin B5, or an aptamer that specifically binds to gluten exorphin B5.

[0033] In some embodiments, the antibody is well known in the art, and the antibody refers to a specific immunoglobulin against an antigenic site. The antibody described in the present invention refers to an antibody that specifically binds to the gluten exorphin B5 described in the present invention, and the antibody that specifically binds to the gluten exorphin B5 can be prepared according to conventional methods in the art.

[0034] Further, the antibody may be in the form of a polyclonal antibody or a monoclonal antibody, an antibody fragment, a single-chain antibody, a multispecific antibody (such as a bispecific antibody), a monospecific antibody, a monovalent antibody, a chimeric antibody, a humanized antibody, a human antibody, a fusion protein comprising an antibody antigen binding site, and any other modified immunoglobulin molecule comprising an antigen binding site, as long as the antibody exhibits the desired biological binding activity.

[0035] In some embodiments, the aptamer refers to a polynucleotide composed of a specific type of single-stranded nucleic acid (DNA, RNA or modified nucleic acid), which has a stable tertiary structure and has the property of being able to bind to the target molecule (gluten exorphin B5 of the present invention) with high affinity and specificity. As described above, since the aptamer can specifically bind to antigenic substances like an antibody, but is more stable than a protein, has a simpler structure, and is composed of a polynucleotide that is easy to synthesize, it can be used instead of an antibody.

[0036] A sixth aspect of the present invention provides a diagnostic product for diagnosing hyperthyroidism-related arrhythmia.

[0037] Furthermore, the diagnostic product comprises a reagent for detecting the content of gluten exorphin B5 in a sample.

[0038] Furthermore, the reagents include reagents for detecting the content of gluten exorphin B5 in a sample by mass spectrometry, chromatography, liquid chromatography-mass spectrometry, gas chromatography-mass spectrometry, enzyme-linked immunosorbent assay, fluorescence, nuclear magnetic resonance or spectroscopy.

[0039] Furthermore, the diagnostic product includes a detection kit, a detection chip or a detection test strip.

[0040] Furthermore, the sample includes a blood sample, a tissue sample, a serum sample, a plasma sample and / or a cell sample.

[0041] In some embodiments, the test kit also includes instructions for use or labels, positive controls, negative controls, buffers, adjuvants or solvents, and one or more containers for containing the composition contained in the test kit. The composition can be in liquid form or can be lyophilized. Suitable containers for the composition include, for example, bottles, vials, syringes and test tubes. The container can be formed of a variety of materials, including glass or plastic. The instructions or labels describe in detail how to use the test kit to test the sample and how the test kit is used to diagnose hyperthyroid arrhythmias.

[0042] In some embodiments, the detection kit may also include a variety of different reagents suitable for practical use (such as for different detection methods), and is not limited to the reagents currently listed in the present invention. As long as the reagent is based on the detection of gluten exorphin B5 to diagnose or assist in the diagnosis of hyperthyroid arrhythmia, it falls within the protection scope of the present invention.

[0043] In some embodiments, the diagnostic kit includes, but is not limited to, an ELISA kit, an immunoblotting detection kit, an immunochromatography detection kit, a flow cytometry analysis kit, an immunohistochemistry detection kit, and an electrochemiluminescence detection kit.

[0044] In some embodiments, the preparation of the detection chip can adopt the conventional preparation method of biochip known to those skilled in the art, including but not limited to: using a solid phase carrier of a modified glass slide or silicon wafer, the 5' end of the probe contains an amino-modified poly dT string, the oligonucleotide probe is prepared into a solution, and then it is spotted on the modified glass slide or silicon wafer using a spotter, arranged into a predetermined sequence or array, and then fixed by leaving it overnight to obtain the detection chip of the present invention.

[0045] The present invention also provides the use of gluten exorphin B5 in constructing a hyperthyroidism arrhythmia early warning model.

[0046] Furthermore, the hyperthyroidism-induced arrhythmia early warning model uses the content of gluten exorphin B5 in the sample as an input variable.

[0047] The present invention also provides a hyperthyroidism arrhythmia early warning system / device based on gluten exorphin B5.

[0048] Further, the system / device comprises:

[0049] A model loading module, used to load the hyperthyroidism arrhythmia early warning model constructed in the application of the present invention as described above;

[0050] An index value acquisition module, used for acquiring the corresponding index value of gluten exorphin B5 in the sample of the subject to be tested according to the hyperthyroidism arrhythmia early warning model;

[0051] An index value detection module is used to determine whether the index value of the corresponding gluten exorphin B5 in the sample of the tested subject is lower than a preset normal value range;

[0052] The hyperthyroidism arrhythmia warning module is used to output the hyperthyroidism arrhythmia warning information of the subject to be tested according to the judgment result.

[0053] In some embodiments, the receiver operating characteristic (ROC) analysis method is used to evaluate the diagnostic performance (i.e., diagnostic efficacy) of the gluten exorphin B5 for the disease of hyperthyroidism arrhythmia, or to test its accuracy in separating various groups of people (in the present invention, especially the group of patients with hyperthyroidism arrhythmia and the healthy control group).

[0054] An ROC curve is a graphical representation of a spectrum of sensitivity and specificity generated using sensitivity as the y-axis and 1-specificity as the x-axis using various cutoff values. In an ROC curve, the true positive rate (sensitivity) is plotted as a function of the FP rate (100-specificity) for different cutoff points. Each point on the ROC curve represents a sensitivity / specificity pair corresponding to a particular decision threshold. The ROC curve for a test with perfect discrimination (no overlap in the two distributions) passes through the upper left corner (sensitivity of 100% and specificity of 100%). Therefore, qualitatively speaking, the closer the plot is to the upper left corner, the better the overall accuracy of the test.

[0055] The area under the ROC curve (AUC) reflects the accuracy of the test and is displayed in the lower left corner of the curve. When AUC is greater than 0.5, the closer it is to 1, the better the diagnostic or predictive effect of the indicator. When AUC is between 0.5-0.7, it has lower accuracy, when AUC is between 0.7-0.9, it has certain accuracy, and when AUC is above 0.9, it has higher accuracy.

[0056] The present invention has been verified to find that the marker gluten exorphin B5, which is significantly differentially expressed between patients with hyperthyroid arrhythmia and healthy controls, has an AUC value of up to 0.9761 when used to diagnose hyperthyroid arrhythmia, indicating that the marker gluten exorphin B5 provided by the present invention can be used for the effective and accurate diagnosis of hyperthyroid arrhythmia.

[0057] The beneficial effects of the present invention are as follows:

[0058] The present invention discovered for the first time the correlation between gluten exorphin B5 and the diagnosis of hyperthyroid arrhythmia, and verified it in a clinically collected sample set. The gluten exorphin B5 has extremely high diagnostic efficacy for diagnosing hyperthyroid arrhythmia, with high accuracy, sensitivity and specificity, an AUC value of up to 0.9761, a sensitivity of up to 94.92%, and a specificity of up to 95.92%. Based on this, the present invention has pioneered the development of an effective prediction method for hyperthyroid arrhythmia, and provided a related prediction system, computer equipment, and computer-readable storage medium, providing information for early intervention treatment of hyperthyroid arrhythmia, and has broad application prospects. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0060] Figure 1 A schematic flow chart of a method for predicting hyperthyroidism-related arrhythmia provided by an embodiment of the present invention;

[0061] Figure 2 A schematic diagram of a prediction system for hyperthyroidism-related arrhythmia provided by an embodiment of the present invention;

[0062] Figure 3 A schematic diagram of a computer device provided in an embodiment of the present invention;

[0063] Figure 4 This is the result graph corresponding to the significant differential expression of gluten exorphin B5 between patients with hyperthyroidism and arrhythmia and healthy controls in the screening set;

[0064] Figure 5 This is a graph showing the diagnostic efficacy of gluten exorphin B5 for diagnosing hyperthyroidism-related arrhythmia in the screening set;

[0065] Figure 6 This is the result graph corresponding to the significant differential expression of gluten exorphin B5 between patients with hyperthyroidism and arrhythmia and healthy controls in the validation set;

[0066] Figure 7 This is a graph showing the diagnostic performance of gluten exorphin B5 for diagnosing hyperthyroidism-related arrhythmia in the validation set. DETAILED DESCRIPTION

[0067] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0068] In some of the processes described in the specification and claims of the present invention and the above-mentioned drawings, multiple operations appearing in a specific order are included, but it should be clearly understood that these operations may not be executed in the order in which they appear in this document or may be executed in parallel. The serial numbers of the operations, such as S101, S102, etc., are only used to distinguish different operations, and the serial numbers themselves do not represent any execution order. In addition, these processes may include more or fewer operations, and these operations may be executed in sequence or in parallel.

[0069] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0070] Figure 1 A flow chart of a method for predicting hyperthyroidism arrhythmia provided in an embodiment of the present invention is provided. Specifically, the method comprises the following steps:

[0071] S101: Obtaining metabolite content data of the sample to be tested;

[0072] In one embodiment, the sample to be tested refers to a sample to be tested derived from a subject to be tested, which can be obtained from the subject's blood and other fluid samples and tissue samples of biological origin, such as a biopsy tissue sample or a tissue culture or cell derived therefrom. The source of the tissue sample can be a solid tissue, such as a fresh, frozen and / or preserved organ or tissue sample, a biopsy tissue or an aspirate; blood or any blood component; body fluid; cells from any time of individual pregnancy or development; or plasma.

[0073] In one embodiment, the sample includes a biological sample that has been processed in any way after it is obtained, such as treated with reagents, stabilized, or enriched for certain components (such as proteins or polynucleotides), or embedded in a semi-solid or solid matrix for sectioning purposes. The subject-derived sample of the present invention includes, but is not limited to: subject-derived blood, tissue, blood-derived cells, serum, plasma, lymph, synovial fluid, cell extracts, and combinations thereof. In a specific embodiment of the present invention, the subject-derived sample is selected from a subject-derived blood sample.

[0074] In one embodiment, the subject refers to any animal, including humans and non-human animals. The non-human animals include all vertebrates, for example, mammals, such as non-human primates (particularly higher primates), sheep, dogs, rodents (such as mice or rats), guinea pigs, goats, pigs, cats, rabbits, cattle, and any livestock or pets; and non-mammals, such as chickens, amphibians, reptiles, etc. In a specific embodiment of the present invention, the subject is a human.

[0075] In one embodiment, the metabolite content data is metabolite content data obtained by detecting the sample to be tested by mass spectrometry, chromatography, liquid chromatography-mass spectrometry, gas chromatography-mass spectrometry, enzyme-linked immunosorbent assay, fluorescence, nuclear magnetic resonance or spectroscopy.

[0076] In one embodiment, the mass spectrometry includes but is not limited to: matrix-assisted activation analysis / ionization time-of-flight mass spectrometer (MALDI-TOF MS), electron spray ionization mass spectrometry (ESI), quadrupole mass spectrometer, ion trap mass spectrometer, MALDI-TOF-TOF mass spectrometry, electrospray ionization (ESI)-TOF mass spectrometry, MALDI quadrupole-time of flight (Q-TOF) mass spectrometry, ESI-TOF-TOF, ESI-Q-TOF, ESI triple quadrupole mass spectrometry, ESI-ion trap mass spectrometry, ESI Fourier transform mass spectrometry (FTMS), MALDI-ion trap-TOF and ESI-ion trap TOF, MALDI-FT MS. Mass spectrometry involves ionizing molecules and subsequently measuring the mass of the resulting ions. Since molecules are ionized in a known manner, the molecular weight of the molecule can be accurately determined by the mass of the ions.

[0077] In one embodiment, the chromatography includes, but is not limited to, high performance liquid chromatography (HPLC), ultra high performance liquid chromatography (UHPLC), gas chromatography (GC), liquid chromatography (LC), and capillary electrophoresis (CE).

[0078] In one embodiment, the spectroscopy includes but is not limited to: near infrared spectroscopy (NIR), ultraviolet visible spectroscopy (UV-Vis), infrared spectroscopy (IR), and nuclear magnetic resonance spectroscopy.

[0079] In a specific embodiment of the present invention, non-targeted metabolomics sequencing, liquid chromatography-mass spectrometry (LCMS) technology and enzyme-linked immunosorbent assay (ELISA) technology are used to detect the metabolite content in the test sample derived from the subject.

[0080] S102: extracting content data of a target metabolite from the metabolite content data, wherein the target metabolite is gluten exorphin B5;

[0081] In one embodiment, the present invention collected 89 blood samples from patients with hyperthyroid arrhythmia in the Department of Cardiology and Endocrinology of the 416th Hospital of Nuclear Industry, and 87 blood samples from healthy controls for subsequent detection using non-targeted metabolomics sequencing (liquid chromatography mass spectrometry (LCMS) technology). The inclusion and exclusion criteria for patients with hyperthyroid arrhythmia are as follows:

[0082] Inclusion criteria: (1) Confirmed hyperthyroidism: including thyroid hormone (FT3, FT4) levels significantly higher than the normal range, and thyroid stimulating hormone (TSH) levels lower than the normal range. (2) Obvious cardiac symptoms: patients have clinical manifestations of cardiac symptoms (arrhythmias) caused by hyperthyroidism. (3) Age and gender: no restrictions, but must be clearly recorded because this may affect disease manifestations and treatment response. (4) Consent to treatment: patients must clearly consent to receive relevant treatment and follow-up.

[0083] Exclusion criteria: (1) Other types of heart disease: such as coronary artery disease, congenital heart disease, etc., which may not be related to hyperthyroidism. (2) Major comorbidities: such as severe liver and kidney dysfunction, uncontrolled hypertension or diabetes, these conditions may interfere with the treatment effect or increase the risk of treatment. (3) Pregnancy: because hormonal changes during pregnancy may affect the interpretation of thyroid function tests, and some drugs for treating hyperthyroidism may be harmful to the fetus. (4) History of drug use: currently using drugs that may affect thyroid function, such as iodine, lithium salts, etc. (5) Refusal to participate: patients refuse to sign informed consent or are unwilling to follow the treatment plan. (6) People with large differences in dietary habits.

[0084] The above samples were randomly divided into a screening set and a validation set. In the screening set, the number of hyperthyroid arrhythmia samples: healthy control samples = 30 cases: 38 cases, and in the validation set, the number of hyperthyroid arrhythmia samples: healthy control samples = 59 cases: 49 cases.

[0085] In one embodiment, during the screening stage, samples are first collected. The disease samples are 30 blood samples from patients with hyperthyroid arrhythmia collected from the Department of Cardiology and the Department of Endocrinology of the 416th Nuclear Industry Hospital in the above screening center, and the control samples are 38 blood samples from healthy controls collected during the previous study in the above screening center.

[0086] Then, the sample is sequenced, and the specific sequencing method is as follows: conventional non-targeted metabolomics sequencing in the art and liquid chromatography-mass spectrometry (LCMS) technology are used to sequence the blood sample. The specific sequencing steps are as follows:

[0087] (1) Sample processing

[0088] Blood sample collection: About 5-10 ml of blood was collected from the subject's peripheral vein using a blood collection tube containing an anticoagulant (such as EDTA or heparin). After sample collection, the sample was quickly centrifuged (4°C, 1500 g, 10 minutes), plasma was separated and immediately stored in a -80°C refrigerator to prevent metabolite degradation.

[0089] Plasma metabolite extraction: Ice-cold methanol-acetonitrile-water (ratio 2:2:1) was used as the extraction solvent to achieve protein precipitation and metabolite extraction. The specific steps are as follows: add 400 µL of extraction solvent to 100 µL of plasma sample; vortex mix for 15 seconds, and ultrasonicate in an ice bath at 4°C for 15 minutes to promote metabolite release; centrifuge at 13,000 g for 15 minutes at 4°C to collect the supernatant; transfer the supernatant to a new tube and dry it in a vacuum centrifuge; resuspend the dried material in an appropriate amount of 75 µL of 50% methanol-water solution, mix thoroughly and set aside.

[0090] Sample cleanup: If necessary, a solid phase extraction (SPE) step can be performed using a C18 column (Waters Sep-Pak C18) to remove interfering substances. The resuspended sample is passed through the equilibrated SPE column, the effluent is collected, dried, and then resuspended for LC-MS analysis.

[0091] (2) Liquid chromatography-mass spectrometry (LC-MS) analysis

[0092] Liquid chromatography (LC) conditions: A reverse phase column (ACQUITY UPLC HSS T3, 2.1 x 100 mm, 1.8 µm, Waters) was used to separate polar metabolites.

[0093] Mobile phase: A two-phase system is usually used: mobile phase A: 0.1% formic acid aqueous solution, mobile phase B: 0.1% formic acid acetonitrile solution.

[0094] Gradient elution program: Depending on the complexity of the sample and the polarity of the target metabolites, the common gradient elution program is: initial: 95% A, 5% B; 0-2 minutes, maintain initial conditions; 2-15 minutes, linearly increase B to 95%; 15-18 minutes, maintain 95% B; 18-20 minutes, return to initial conditions and equilibrate the column.

[0095] Flow rate: Typically 0.3 mL / min.

[0096] Injection volume: Typically 2-5 µL per injection.

[0097] Mass spectrometry (MS) conditions: High-resolution mass spectrometry (Thermo Scientific Q Exactive or Agilent 6545 Q-TOF) was used.

[0098] Ionization mode: Depending on the nature of the metabolites, electrospray ionization (ESI) was used, which can scan in both positive and negative ion modes simultaneously.

[0099] Mass spectrometry scan range: usually set to m / z 70-1000 to cover small molecule metabolites.

[0100] Resolution: High-resolution mode (typically 70,000 or higher) ensures accurate mass-to-charge ratio (m / z) identification of metabolites.

[0101] Fragmentation mode: Data-dependent acquisition (DDA) mode was used to obtain secondary MS / MS information of metabolites.

[0102] Collision energy: Multi-step collision energy (20, 40, 60 eV) was set to ensure effective fragmentation of different metabolites.

[0103] (3) Data processing

[0104] Data processing includes mass spectrometry data preprocessing, database matching, metabolite identification, and metabolite quantification. The specific data processing methods are as follows:

[0105] Mass spectrometry data preprocessing

[0106] Data denoising: A dedicated data processing software (Thermo Xcalibur) was used to perform peak identification, denoising, and baseline correction on the raw mass spectrometry data.

[0107] Normalization: In order to reduce the systematic error between samples, the internal standard method or total ion current (TIC) method is often used for normalization.

[0108] Internal standard method: Add known concentrations of internal standard metabolites (13C-labeled metabolites) before sample extraction for quantitative correction.

[0109] TIC method: The peak area of ​​each metabolite was normalized according to the total ion flow of the sample.

[0110] Database matching and metabolite identification

[0111] Database matching: Metabolites are identified by comparing with existing metabolite mass spectrometry databases, such as HMDB (Human Metabolome Database), METLIN, MassBank, etc. Identification is usually based on the mass-to-charge ratio (m / z) of the primary mass spectrometry and the fragment information of the secondary mass spectrometry.

[0112] Pattern recognition: Combined with multivariate statistical analysis methods (principal component analysis PCA, partial least squares discriminant analysis PLS-DA), metabolome data were analyzed to reveal the differences in metabolites between different sample groups.

[0113] Metabolite quantification

[0114] Absolute quantification: If absolute quantification of a specific metabolite is required, a standard curve can be drawn using a standard substance of known concentration to calculate the metabolite concentration in the sample.

[0115] Relative quantification: For metabolites without standards, relative quantification was performed using peak area, and inter-group comparison was performed using normalized data.

[0116] (4) Interpretation and analysis of results

[0117] Data statistical analysis: SPSS, R language or MetaboAnalyst and other statistical software were used to perform statistical analysis on the quantitative results. The t-test, ANOVA and other methods were used to test the differences between the groups and determine the differential metabolites. The adjusted P value or FDR was used to control the multiple testing error.

[0118] Biological pathway analysis: Based on the identification results of metabolites, biological pathway databases such as KEGG and Reactome are used to analyze the position of differential metabolites in biological metabolic pathways and reveal their role in biological functions.

[0119] In one embodiment, during the verification stage, samples are first collected. The disease samples are 59 blood samples from patients with hyperthyroid arrhythmia collected from the Department of Cardiology and the Department of Endocrinology of the 416th Nuclear Industry Hospital in the above verification set, and the control samples are 49 blood samples from healthy controls collected in the previous study in the above verification set.

[0120] Then, the sample is tested. The specific testing method is as follows: the gluten exorphin B5 level in the blood sample is tested using a conventional ELISA testing method in the art. The specific testing steps are as follows:

[0121] (1) Coating: Dilute the coating antigen to 5 μg / mL and use CBS coating solution at pH 9.6. Evenly add the diluted solution to each well of the ELISA plate, 50 μL per well. Perform coating at room temperature overnight. (2) Washing: Pour out the coating solution completely and ensure that the wells are dry. Wash the plate three times with 1×PBST washing solution at pH 7.2-7.4, making sure that each well is filled each time, and continue for 3 min-5 min. (3) Blocking: Add 250 μL of blocking solution (1% BSA) to each well, and then block at 37°C for 2 hours. (4) Re-washing: Repeat the washing step three times with 1×PBST washing solution, each washing lasting 3 min-5 min. (5) Binding to primary antibody: Dilute the blood sample to be tested with 1×PBST buffer at a ratio of 1:1000, and add 100 μL to each well. At the same time, set up a negative control and incubate at 37°C for 1.5 hours. (6) Perform the washing step again: the method is the same as the above step (4). (7) Bind the enzyme-labeled secondary antibody: add 50 μL of the enzyme-labeled secondary antibody goat anti-mouse IgG-HRP to each well, which has been diluted 1000 times with 1× PBST buffer. Incubate at 37°C for 1 hour. (8) Perform the washing step again: the method is the same as the above step (4). (9) Color development: add 50 μL of substrate solution (TMB) to each well, and then color develop at room temperature in the dark for 10 minutes. (10) Stop color development and read the results: after color development is completed, add 50 μL of stop solution (2 mol / L H2SO4) to each well to stop the reaction. Use a microplate reader to read the OD value at 450 nm. If it is greater than 2.1 times the specified negative control OD value, it is positive.

[0122] In one embodiment, the present invention verifies the diagnostic efficacy of the gluten exorphin B5 in a screening set and a validation set.

[0123] First, for the marker gluten exorphin B5 that was screened in this example and was significantly differentially expressed between patients with hyperthyroid arrhythmia and healthy controls, the R package "pROC" (version 1.15.0) was used to draw the receiver operating curve (ROC curve), and its sensitivity, specificity, and AUC value were analyzed to determine its diagnostic efficacy in the screening set and the validation set.

[0124] Among them, when judging the diagnostic efficacy of the marker gluten exorphin B5 in the screening set and the validation set, the content or expression level of gluten exorphin B5 is directly used for analysis, and the level corresponding to the point with the largest Youden index is selected as its cutoff value. Among them, the obtained AUC value is the area enclosed by the coordinate axis under the ROC curve, and its value range is between 0.5 and 1. The larger the AUC value, the higher the accuracy of the marker for diagnosing hyperthyroid arrhythmia; the AUC value close to 1.0 indicates that the authenticity of the diagnostic result obtained based on the marker is high and the prediction effect is better.

[0125] In one embodiment, in the screening set (hyperthyroidism arrhythmia samples: healthy control samples = 30 cases: 38 cases), the results of non-targeted metabolomics analysis are shown in Figure 4 The results showed that compared with healthy controls, the level of gluten exorphin B5 in blood samples of patients with hyperthyroid arrhythmia was significantly reduced, and the difference was statistically significant (P<0.05).

[0126] In the screening set (hyperthyroid arrhythmia samples: healthy control samples = 30 cases: 38 cases), the diagnostic efficacy results of the differentially expressed metabolite gluten exorphin B5 for diagnosing hyperthyroid arrhythmia are shown in Figure 5 The results showed that gluten exorphin B5 has high accuracy, sensitivity and specificity in the diagnosis of hyperthyroidism-induced arrhythmia, among which the AUC value is as high as 0.8886, the sensitivity is 94.74%, and the specificity is 83.33%, indicating that the gluten exorphin B5 can be used in the effective diagnosis of hyperthyroidism-induced arrhythmia.

[0127] In one embodiment, in the validation set (hyperthyroidism arrhythmia samples: healthy control samples = 59 cases: 49 cases), the results of further analysis by ELISA are shown in Figure 6 The results showed that compared with healthy controls, the level of gluten exorphin B5 in blood samples of patients with hyperthyroid arrhythmia was significantly reduced, and the difference was statistically significant (P<0.05).

[0128] In one embodiment, in the validation set (hyperthyroidism arrhythmia samples: healthy control samples = 59 cases: 49 cases), the diagnostic efficacy of the gluten exorphin B5 for diagnosing hyperthyroidism arrhythmia is shown in Figure 7The results showed that gluten exorphin B5 also had high accuracy, sensitivity and specificity in the diagnosis of hyperthyroidism-induced arrhythmia in the validation set, with an AUC value as high as 0.9761, a sensitivity of 94.92%, and a specificity of 95.92%. The above results once again proved that gluten exorphin B5 can be used in the effective diagnosis and prediction of hyperthyroidism-induced arrhythmia.

[0129] S103: performing classification prediction based on the content data of the target metabolite to obtain a classification result of whether the sample to be tested is a hyperthyroidism arrhythmia sample;

[0130] The classification result is obtained based on a prediction model, and the method for constructing the prediction model includes: obtaining the content data of the target metabolite of the training set samples and the clinical characteristics corresponding to the samples, the clinical characteristics including patients with hyperthyroid arrhythmia and healthy people, extracting the content data of the target metabolite in the training set and inputting it into a machine learning model to construct a prediction model, and obtaining a constructed prediction model and threshold;

[0131] If the content of the target metabolite is lower than the threshold value, the classification result of the sample to be tested is a hyperthyroidism arrhythmia sample; if the content of the target metabolite is higher than the threshold value, the classification result of the sample to be tested is a non-hyperthyroidism arrhythmia sample.

[0132] In one embodiment, the machine learning model is a linear regression model, a logistic regression model, a Lasso regression model, a Ridge regression model, a linear discriminant analysis model, a random forest model, a nearest neighbor model, a naive Bayes model, a decision tree model, a perceptron model, a neural network model, a support vector machine model, an AdaBoost model, a GBDT model, an XGBoost model, a LightGBM model or a CatBoost model.

[0133] In one embodiment, the threshold in a machine learning model refers to a critical value used to convert the model output into a classification result. In a binary classification problem, the model usually outputs a real value, which represents the probability of belonging to a certain category. For example, a logistic regression model outputs a probability value between 0 and 1, and the default threshold is usually 0.5, that is, probabilities greater than or equal to 0.5 are classified as positive, and probabilities less than 0.5 are classified as negative.

[0134] In one embodiment, the precision-recall curve can be used to select the best threshold. By plotting the precision-recall curve, the impact of different thresholds on model performance can be intuitively seen. The point on the curve is selected to achieve the best balance between precision and recall, and this point is the best threshold.

[0135] In one embodiment, the optimal threshold can be selected using the F1 score, which is the harmonic mean of precision and recall. The optimal threshold is selected by maximizing the F1 score.

[0136] In one embodiment, the ROC curve can be used to select the best threshold. The receiver operating characteristic curve (ROC) can also be used to select the best threshold, and the performance of the model is evaluated by calculating the area under the curve (AUC) to determine the best threshold.

[0137] In one embodiment, when the above-mentioned prediction model construction method is determined, the prediction model includes the threshold value, that is, when the prediction model is determined, the threshold value is also determined. Based on the determined threshold value, it is possible to predict whether the classification result of the sample to be tested is a sample of congenital heart disease combined with intellectual disability. The specific judgment result based on the prediction model is: if the content of the target metabolite is lower than the threshold value, the classification result of the sample to be tested is a sample of hyperthyroidism arrhythmia; if the content of the target metabolite is higher than the threshold value, the classification result of the sample to be tested is a sample of non-hyperthyroidism arrhythmia.

[0138] In one embodiment, the effectiveness of the constructed prediction model can also be predicted, that is, another data set containing the content data of the target metabolite corresponding to patients with hyperthyroid arrhythmia and healthy people is taken, and the effectiveness of the constructed prediction model is verified in the data set.

[0139] Figure 2 A schematic diagram of a prediction system for hyperthyroidism-induced arrhythmia provided by an embodiment of the present invention, specifically, the system comprises:

[0140] A data acquisition unit, used to acquire metabolite content data of the sample to be tested;

[0141] A data extraction unit, used for extracting content data of a target metabolite from the metabolite content data, wherein the target metabolite is gluten exorphin B5;

[0142] A result prediction unit, used for performing classification prediction based on the content data of the target metabolite to obtain a classification result of whether the sample to be tested is a hyperthyroidism arrhythmia sample;

[0143] The classification result is obtained based on a prediction model, and the method for constructing the prediction model includes: obtaining the content data of the target metabolite of the training set samples and the clinical characteristics corresponding to the samples, the clinical characteristics including patients with hyperthyroid arrhythmia and healthy people, extracting the content data of the target metabolite in the training set and inputting it into a machine learning model to construct a prediction model, and obtaining a constructed prediction model and threshold;

[0144] If the content of the target metabolite is lower than the threshold value, the classification result of the sample to be tested is a hyperthyroidism arrhythmia sample; if the content of the target metabolite is higher than the threshold value, the classification result of the sample to be tested is a non-hyperthyroidism arrhythmia sample.

[0145] In one embodiment, the machine learning model is a linear regression model, a logistic regression model, a Lasso regression model, a Ridge regression model, a linear discriminant analysis model, a random forest model, a nearest neighbor model, a naive Bayes model, a decision tree model, a perceptron model, a neural network model, a support vector machine model, an AdaBoost model, a GBDT model, an XGBoost model, a LightGBM model or a CatBoost model.

[0146] In one embodiment, the metabolite content data is metabolite content data obtained by detecting the sample to be tested by mass spectrometry, chromatography, liquid chromatography-mass spectrometry, gas chromatography-mass spectrometry, enzyme-linked immunosorbent assay, fluorescence, nuclear magnetic resonance or spectroscopy.

[0147] In one embodiment, the sample to be tested is a blood sample, a serum sample, a plasma sample, a tissue sample or a cell sample.

[0148] Figure 3 A schematic diagram of a computer device provided for an embodiment of the present invention, specifically, the computer device includes a memory and a processor, the memory is used to store program instructions; the processor is used to call the program instructions, and when the program instructions are executed, the prediction method for hyperthyroidism-related arrhythmia as described above is implemented.

[0149] An embodiment of the present invention further provides a computer-readable storage medium. Specifically, a computer program is stored on the computer-readable storage medium. When the computer program is executed by a processor, the method for predicting hyperthyroidism-induced arrhythmia as described above is implemented.

[0150] The embodiment of the present invention also provides the use of a reagent for detecting the content of gluten exorphin B5 in a sample to be tested in the preparation of a diagnostic product for diagnosing hyperthyroidism-related arrhythmia.

[0151] The embodiment of the present invention also provides a diagnostic product for diagnosing hyperthyroidism-related arrhythmia, wherein the diagnostic product comprises a reagent for detecting the content of gluten exorphin B5 in a sample.

[0152] The validation results of this validation example show that assigning intrinsic weights to indications can moderately improve the performance of the method relative to the default setting.

[0153] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0154] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0155] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0156] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0157] A person skilled in the art may understand that all or part of the steps in the various methods of the above embodiments may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, and the storage medium may include: a read-only memory (ROM), a random access memory (RAM), a disk or an optical disk, etc.

[0158] A person skilled in the art can understand that all or part of the steps in the above-mentioned embodiment method can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable storage medium. The above-mentioned storage medium can be a read-only memory, a disk or an optical disk, etc.

[0159] The above is a detailed introduction to a computer device provided by the present invention. For a person skilled in the art, according to the concept of the embodiments of the present invention, there may be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as limiting the present invention.

Claims

1. A method for predicting hyperthyroidism-related arrhythmia, characterized in that: The method comprises: Obtaining metabolite content data of the sample to be tested; extracting content data of a target metabolite from the metabolite content data, wherein the target metabolite is gluten exorphin B5; Performing classification prediction based on the content data of the target metabolite to obtain a classification result of whether the sample to be tested is a hyperthyroidism arrhythmia sample; The classification result is obtained based on a prediction model, and the method for constructing the prediction model includes: obtaining the content data of the target metabolite of the training set sample and the clinical characteristics corresponding to the sample, the clinical characteristics including the classification data of whether the subject from which the sample is derived suffers from hyperthyroid arrhythmia, extracting the content data of the target metabolite in the training set and inputting it into a machine learning model to construct a prediction model, and obtaining a constructed prediction model and threshold; If the content of the target metabolite is lower than the threshold value, the classification result of the sample to be tested is a hyperthyroidism arrhythmia sample; if the content of the target metabolite is higher than the threshold value, the classification result of the sample to be tested is a non-hyperthyroidism arrhythmia sample.

2. The method for predicting hyperthyroidism-related arrhythmia according to claim 1, characterized in that: The machine learning model is a linear regression model, a logistic regression model, a Lasso regression model, a Ridge regression model, a linear discriminant analysis model, a random forest model, a nearest neighbor model, a naive Bayes model, a decision tree model, a perceptron model, a neural network model, a support vector machine model, an AdaBoost model, a GBDT model, an XGBoost model, a LightGBM model or a CatBoost model.

3. The method for predicting hyperthyroidism-related arrhythmia according to claim 1, characterized in that: The metabolite content data are metabolite content data obtained by detecting the sample to be tested by mass spectrometry, chromatography, liquid chromatography-mass spectrometry, gas chromatography-mass spectrometry, enzyme-linked immunosorbent assay, fluorescence, nuclear magnetic resonance or spectroscopy.

4. The method for predicting hyperthyroidism-related arrhythmia according to claim 1, characterized in that: The sample to be tested is a blood sample, a serum sample, a plasma sample, a tissue sample or a cell sample.

5. A prediction system for hyperthyroidism-related arrhythmia, characterized in that: The system comprises: A data acquisition unit, used to acquire metabolite content data of the sample to be tested; A data extraction unit, used for extracting content data of a target metabolite from the metabolite content data, wherein the target metabolite is gluten exorphin B5; A result prediction unit, used for performing classification prediction based on the content data of the target metabolite to obtain a classification result of whether the sample to be tested is a hyperthyroidism arrhythmia sample; The classification result is obtained based on a prediction model, and the method for constructing the prediction model includes: obtaining the content data of the target metabolite of the training set sample and the clinical characteristics corresponding to the sample, the clinical characteristics including the classification data of whether the subject from which the sample is derived suffers from hyperthyroid arrhythmia, extracting the content data of the target metabolite in the training set and inputting it into a machine learning model to construct a prediction model, and obtaining a constructed prediction model and threshold; If the content of the target metabolite is lower than the threshold value, the classification result of the sample to be tested is a hyperthyroidism arrhythmia sample; if the content of the target metabolite is higher than the threshold value, the classification result of the sample to be tested is a non-hyperthyroidism arrhythmia sample.

6. The prediction system for hyperthyroidism-induced arrhythmia according to claim 5, characterized in that: The machine learning model is a linear regression model, a logistic regression model, a Lasso regression model, a Ridge regression model, a linear discriminant analysis model, a random forest model, a nearest neighbor model, a naive Bayes model, a decision tree model, a perceptron model, a neural network model, a support vector machine model, an AdaBoost model, a GBDT model, an XGBoost model, a LightGBM model or a CatBoost model.

7. A computer device, characterized in that: The computer device includes a memory and a processor, the memory is used to store program instructions; the processor is used to call the program instructions, and when the program instructions are executed, the method for predicting hyperthyroidism arrhythmia described in any one of claims 1-4 is implemented.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method for predicting hyperthyroidism arrhythmia according to any one of claims 1 to 4 is implemented.

Images