Probes for detecting SNPs related to venous thrombosis and their applications and venous thrombosis risk prediction system

Abstract

The present invention relates to the field of gene engineering technology, and in particular to a probe for detecting SNPs related to venous thrombosis, its application and a venous thrombosis risk prediction system. The present invention designs and synthesizes polycyclic aromatic bases with specific structures and substituents, and modifies them onto triplex DNA, thereby enhancing the binding force between triplex DNA and double-stranded DNA and the ability to recognize SNPs. The modified triplex DNA forms a stable structure with the double-stranded DNA, thereby specifically recognizing the target SNP. When the triplex DNA binds to the target SNP, the fluorescence wavelength of the fluorescent marker moves, thereby realizing the detection of the SNP, and using machine learning to analyze the detection data to predict the risk of venous thrombosis. The present invention improves the accuracy and specificity of venous thrombosis risk prediction, and provides strong support for early prevention and intervention of the disease.

Description

Technical Field

[0001] The present invention relates to the technical field of gene detection, and in particular to a probe for detecting SNPs related to venous thrombosis and its application and a venous thrombosis risk prediction system. Background Art

[0002] Venous thrombosis is a common and serious vascular disease, which is mainly caused by abnormal coagulation of blood in the veins, leading to vascular obstruction. Venous thrombosis can occur in any part of the body, but deep vein thrombosis of the lower extremities and pulmonary embolism are the most common, with high morbidity and mortality. According to statistics, about 10 million people are affected by venous thrombosis every year worldwide, of which about 30% of patients will experience complications of varying degrees, seriously affecting the patient's quality of life and prognosis.

[0003] At present, the commonly used methods for predicting the risk of venous thrombosis in clinical practice mainly include clinical assessment models and laboratory test indicators. Clinical assessment models such as the Wells score and the Caprini score predict the risk of venous thrombosis in patients by comprehensively evaluating the patient's clinical symptoms, signs, medical history and other factors. However, these clinical assessment models have certain limitations, and their accuracy and specificity are affected by many factors, such as individual differences in patients and the complexity of the disease. Laboratory test indicators mainly include D-dimers, coagulation factors, etc. Although these indicators can reflect the patient's coagulation function status to a certain extent, they are still insufficient for the early diagnosis and risk prediction of venous thrombosis.

[0004] With the continuous development of molecular biology technology, the risk prediction method of venous thrombosis based on genetic testing has gradually become a research hotspot. Although traditional genetic testing methods such as polymerase chain reaction (PCR) and restriction fragment length polymorphism (RFLP) can detect gene SNPs, they have problems such as complex operation, high testing cost, and low throughput, making them difficult to be widely used in clinical practice. Summary of the invention

[0005] In view of the above-mentioned deficiencies of the prior art, the present invention aims to provide a probe and application for venous thrombosis-related SNP detection and a venous thrombosis risk prediction system, so as to achieve low-cost and accurate prediction of venous thrombosis risk.

[0006] In order to solve the above problems, the present invention adopts the following technical solutions:

[0007] In the first aspect, the present invention provides a probe for detecting SNPs associated with venous thrombosis. After fluorescently labeled polycyclic aromatic bases are modified onto Triplex DNA designed based on target SNPs, the probe is combined with silicon-based nanomaterials to form a probe; the target SNPs include rs6025, rs8176719, rs1799963, rs2066865, rs2036914, rs1799889, rs5918 and rs1801133.

[0008] As an implementation method, the polycyclic aromatic base is obtained by introducing hydroxyl, amino or methyl groups at the base positions of naphthyl polycyclic aromatic bases, anthracene polycyclic aromatic bases, phenanthryl polycyclic aromatic bases, pyrene polycyclic aromatic bases, fluoranthene polycyclic aromatic bases or benzanthracene polycyclic aromatic bases.

[0009] As an implementation method, the polycyclic aromatic base is a three-membered ring structure.

[0010] As an implementation method, the number of polycyclic aromatic bases modified with fluorescently labeled polycyclic aromatic bases on the Triplex DNA designed by the target SNP is 1 or 2; when the number of the polycyclic aromatic bases is 1, the length of the Triplex DNA is 25bp; when the number of the polycyclic aromatic bases is 2, the length of the Triplex DNA is 15bp.

[0011] As an implementable embodiment, the polycyclic aromatic base is located 2-3 bp from the 5' end of the target SNP, and the interval between the multiple polycyclic aromatic bases is greater than or equal to 3 bp.

[0012] As an implementation method, the nucleotide sequence of the Triplex DNA is shown as SEQ ID NO.1-SEQ ID NO.8.

[0013] In a second aspect, the present invention provides a risk prediction system for venous thrombosis, including a data acquisition module, a model building module, a SNP detection module and a prediction module;

[0014] The data acquisition module is used to collect the target SNP data and clinical characteristic data of patients with venous thrombosis as a training set;

[0015] The model building module is used to train the venous thrombosis risk prediction model based on the obtained training set using a random forest algorithm to learn the relationship between the target SNP test results and clinical feature data and the venous thrombosis risk;

[0016] The SNP detection module is used to detect whether the subject has a target SNP based on the probe for venous thrombosis-related SNP detection using wavelength shift data of a fluorescent marker, and obtain a target SNP detection result of the subject;

[0017] The prediction module is used to input the obtained target SNP detection results and clinical characteristic data of the subject into the venous thrombosis risk prediction model to obtain a venous thrombosis risk score.

[0018] As an implementable embodiment, the clinical characteristic data include age, gender, weight, history of hypertension and history of diabetes.

[0019] As an implementable embodiment, the minimum concentration of the target SNP in the test sample of the subject is 0.5 pM.

[0020] In a third aspect, the present invention provides use of the probe for detecting venous thrombosis-related SNPs in the preparation of a venous thrombosis detection reagent.

[0021] The beneficial effects of the present invention are:

[0022] 1. By designing and synthesizing polycyclic aromatic bases with specific structures and substituents and modifying them onto Triplex DNA, the binding force between Triplex DNA and double-stranded DNA and the ability to recognize SNPs are significantly enhanced, which enables the detection system to more accurately identify target SNPs and reduce the occurrence of false positive and false negative results.

[0023] 2. Fluorescently labeled polycyclic aromatic base-modified Triplex DNA is combined with silicon-based nanomaterials to construct detection probes. Silicon-based nanomaterials have good biocompatibility, stability and optical properties, which can improve the intensity and specificity of the detection signal, which helps to improve the sensitivity of detection, so that SNPs can be accurately detected even in low-concentration samples.

[0024] 3. Use machine learning and neural network models to analyze the SNP test results and clinical information in the patient's genomic DNA to establish a venous thrombosis risk prediction model. This multi-factor comprehensive analysis method can more comprehensively assess the patient's risk of venous thrombosis and provide a scientific basis for early prevention and intervention. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 A comparison chart of the structures of different polycyclic aromatic bases.

[0026] Figure 2 A comparison chart of the number of different polycyclic aromatic bases.

[0027] Figure 3 This is a comparison chart of Triplex DNA of different lengths when the number of polycyclic aromatic bases is 0.

[0028] Figure 4 This is a comparison chart of Triplex DNA of different lengths when the number of polycyclic aromatic bases is 1.

[0029] Figure 5 This is a comparison chart of Triplex DNA of different lengths when the number of polycyclic aromatic bases is 2.

[0030] Figure 6 This is a comparison chart of Triplex DNA of different lengths when the number of polycyclic aromatic bases is 3.

[0031] Figure 7 This is a test chart for detecting the accuracy of the probe.

[0032] Figure 8 This is the minimum detection limit test diagram of the detection probe. DETAILED DESCRIPTION

[0033] The present invention is further described in detail below in conjunction with specific embodiments.

[0034] It should be noted that these embodiments are only used to illustrate the present invention rather than to limit the present invention. Simple improvements to the method based on the concept of the present invention all fall within the scope of protection claimed by the present invention.

[0035] The present invention adopts SNPs of genes such as F5 gene (rs6025), ABO gene (rs8176719), F2 gene (rs1799963), FGG gene (rs2066865), F11 gene (rs2036914), PAI-1 gene (rs1799889), ITGB3 gene (rs5918) and MTHFR gene (rs1801133), and predicts the risk of venous thrombosis by detecting the SNPs of these genes.

[0036] The triplex DNA used in the present invention is a triple-stranded structure formed by a single-stranded DNA and a double-stranded DNA through Hoogsteen hydrogen bonds, which has high specificity and affinity. By designing a specific triplex DNA, specific identification of the target gene SNP is achieved. The development of nanotechnology has provided a new platform for biological detection. Nanomaterials have the advantages of large specific surface area, high surface activity, unique optical properties, etc., which can improve the sensitivity and specificity of detection.

[0037] Example 1: Selection and design of polycyclic aromatic bases

[0038] (1) Comparison of structures of different polycyclic aromatic bases

[0039] In order to investigate the effect of different base types on the stability of Triplex DNA formation, 6 types of polycyclic aromatic bases were designed. The design principles are as follows: a larger planar structure can increase the π-π stacking effect with double-stranded DNA and improve the binding force; the introduction of different substituents can adjust the electronic properties and hydrogen bond formation ability of the bases and enhance the specific recognition of SNPs; considering the stability and synthesis difficulty of polycyclic aromatic bases, a structure that is easy to synthesize and stable under physiological conditions is selected. The specific structures are shown in Table 1.

[0040] Table 1. Structures of different polycyclic aromatic bases

[0041] Using a fluorescence analyzer, the binding of Triplex DNA modified with different polycyclic aromatic bases and double-stranded DNA marked with fluorescent signals is measured. Double-stranded DNA containing the target SNP is fixed on the sensor chip, and then different Triplex DNAs are allowed to flow over the chip surface. The difference in the ability to detect SNPs is determined by detecting the change in fluorescence wavelength. The results are shown in Figure 1. Figure 1 As shown, as the number of benzene rings increases, the binding force of Triplex DNA becomes stronger and the signal value becomes higher, but the change in fluorescence wavelength becomes smaller and the specificity of SNP becomes lower. Therefore, the three-membered rings such as anthracene and phenanthrene are the best choices.

[0042] (2) Comparison of the number of different polycyclic aromatic bases

[0043] In order to investigate the effect of different numbers of polycyclic aromatic bases on the stability of triplex DNA formation, probes with 1-3 anthracene polycyclic aromatic bases were designed. The fluorescence analyzer was also used to measure the binding of triplex DNA modified with different numbers of anthracene polycyclic aromatic bases labeled with fluorescent signals to double-stranded DNA. Figure 2 As shown, as the number of anthracene polycyclic aromatic bases increases, the binding force of Triplex DNA becomes stronger and the signal value becomes higher, but the change in fluorescence wavelength becomes smaller and the specificity of SNP becomes lower.

[0044] (3) Comparison of different probe lengths

[0045] In order to investigate the effect of different probe lengths on the stability of Triplex DNA formation, probes of different lengths were designed with 1-3 anthracene polycyclic aromatic bases incorporated, and the fluorescence analyzer was also used for measurement. Figure 3-6As shown in the figure, when the number of anthracene polycyclic aromatic bases is 0, the optimal length of Triplex DNA is 35bp; when the number of anthracene polycyclic aromatic bases is 1, the optimal length of Triplex DNA is 25bp; when the number of anthracene polycyclic aromatic bases is 2, the optimal length of Triplex DNA is 15bp; when the number of anthracene polycyclic aromatic bases is 3, the specificity of Triplex DNA in detecting SNPs is not good. Therefore, the best effect is to select a Triplex DNA probe with a length of about 15bp containing 2 anthracene polycyclic aromatic bases.

[0046] (4) Comparison of the positions of different polycyclic aromatic bases

[0047] In order to investigate the effect of different probe lengths on the stability of Triplex DNA formation, two anthracene polycyclic aromatic bases were designed to be incorporated into a 15 bp probe, with the SNP position as the coordinate 0 point, the 3' end as a negative value, and the 5' end as a positive value, and the results of different positions were compared using a fluorescence analyzer. The results are shown in Table 2. When the anthracene polycyclic aromatic bases are all located at the 5' end of the SNP site, the effect is better than the 3' end; among them, when the anthracene polycyclic aromatic base is located 2-3 bp from the 5' end of the SNP site, the effect is best, and when the interval between the two anthracene polycyclic aromatic bases is ≥3 bp, the effect is best.

[0048] Table 2. Comparison of the positions of different polycyclic aromatic bases

[0049] Example 2: Selection and activation of silicon-based nanomaterials

[0050] Silicon nanoparticles have good biocompatibility, stability and easy surface modification. They can remain stable in physiological environments, and the surface of silicon nanoparticles can be chemically modified to introduce various functional groups to enhance the ability to bind with other molecules.

[0051] The specific preparation and activation process of silicon nanoparticles is as follows:

[0052] 1. Mix tetraethyl orthosilicate (TEOS, concentration is 0.1 mol / L), ethanol, water and ammonia water in a certain proportion, stir and react at 30°C for 8 hours to obtain silicon nanoparticles with an average particle size of about 50-80 nanometers.

[0053] 2. Use silane coupling agent to activate the surface of silicon nanoparticles. Disperse silicon nanoparticles in ethanol, add a certain amount of silane coupling agent (such as 3-aminopropyltriethoxysilane, APTES), 1-5 ml of APTES per gram of silicon nanoparticles, and stir the reaction at 40-60°C for 2-6 hours.

[0054] Example 3: Binding of Triplex DNA modified with fluorescently labeled polycyclic aromatic bases to silicon-based nanomaterials

[0055] (1) Prepare materials:

[0056] Fluorescently labeled polycyclic aromatic base-modified Triplex DNA: designed and synthesized according to the SNP region of the target gene, labeled with a suitable fluorescent dye to ensure high labeling efficiency without affecting the structure and function of the Triplex DNA, see Table 3 for details.

[0057] Activated silicon-based nanomaterials: silicon nanoparticles prepared and activated according to the method described above.

[0058] Table 3. Fluorescently labeled polycyclic aromatic base-modified Triplex DNA

[0059] Note: X is anthracene polycyclic aromatic base; underline is SNP site

[0060] (2) Experimental steps:

[0061] A. Dissolve the fluorescently labeled Triplex DNA in 0.01 M phosphate buffered saline (PBS, pH 7.4) to a final concentration of 100 nM.

[0062] B. The activated silicon nanoparticles were dispersed in 0.0.1 M phosphate buffer (PBS, pH 7.4) to a final concentration of 1 mg / mL.

[0063] C. Mix the solution containing Triplex DNA and the solution containing silicon nanoparticles in a volume ratio of 1:1 and react at 37°C with magnetic stirring at 200 rpm for 4 hours, usually at 30 - 40°C.

[0064] D. Centrifuge at 14000 rpm for 20 min, discard the supernatant, and add 0.01 M phosphate buffer (PBS, pH 7.4) to a final concentration of 1 mg / mL.

[0065] E. Repeat the previous step and store at 2-8℃ for later use.

[0066] Example 4: Performance test of detection probe

[0067] (1) Accuracy test

[0068] Mix 10 μL of the detection probe solution with the double-stranded DNA sample containing the target SNP and incubate at 60°C for 30 minutes. Use a fluorescence analyzer to measure the wavelength and intensity of the fluorescence signal of the mixed solution. The results are shown in Figure 7 As shown, the Triplex DNA probe can accurately genotype all eight venous thrombosis risk gene SNPs.

[0069] (2) Detection limit detection

[0070] Prepare a series of samples containing different concentrations of the target SNP (e.g., 0.1 pM, 0.5 pM, 1 pM, 5 pM, 10 pM). Mix the detection probe with these samples respectively and measure the fluorescence signal as described above. Evaluate the specificity of the detection probe by comparing the changes in fluorescence intensity of the target SNP at different concentrations. The results are shown in Figure 2. Figure 8 As shown, the Triplex DNA probe can detect SNP samples as low as 0.5 pM.

[0071] Example 5: Establishment of a venous thrombosis risk prediction model

[0072] (1) Data collection

[0073] The genomic SNP test results and clinical information of 518 patients were collected, including the following:

[0074] SNP detection results: SNP detection was performed on F5 gene (rs6025), ABO gene (rs8176719), F2 gene (rs1799963), FGG gene (rs2066865), F11 gene (rs2036914), PAI-1 gene (rs1799889), ITGB3 gene (rs5918) and MTHFR gene (rs1801133), and the results were divided into the presence of specific SNP (denoted as 1) and absence (denoted as 0).

[0075] Clinical information: including age, gender, weight, medical history (whether there is hypertension, diabetes, etc., if yes, record as 1, if no, record as 0).

[0076] (2) Data preprocessing

[0077] A. Data cleaning: Check whether there are missing values, outliers, etc. in the data. Missing values ​​can be processed by methods such as mean filling and median filling; outliers can be deleted or corrected according to actual conditions.

[0078] B. Data coding: Encode categorical variables such as gender, encoding male as 0 and female as 1.

[0079] (3) Feature selection

[0080] The correlation coefficient between each SNP and the occurrence of venous thrombosis was calculated, and the SNP with higher correlation was selected as the feature.

[0081] (4) Model selection and training

[0082] Select appropriate machine learning algorithms and neural network models, such as random forest, support vector machine, convolutional neural network, etc. Here we take random forest as an example:

[0083] A. Determine model parameters: Set the number of trees in the random forest to 100 and the maximum depth to 10.

[0084] B. Divide the preprocessed dataset into training set and test set with a ratio of 7:3.

[0085] C. Use the training set to train the random forest model and let the model learn the relationship between SNP test results and clinical information and venous thrombosis risk.

[0086] (5) Model evaluation and optimization

[0087] The trained model is evaluated using the test set, with the evaluation indicators being accuracy and recall. The accuracy of the calculated model on the test set is set to 85%, and the recall is set to 80%.

[0088] If the model performance is not ideal, you can optimize it by adjusting model parameters, increasing the amount of data, performing feature engineering, etc. For example, increase the number of trees or adjust the maximum depth, retrain the model and evaluate the performance.

[0089] Example 6: Clinical validation of a venous thrombosis risk prediction model

[0090] (1) Clinical data collection and testing

[0091] 107 patient samples were selected to detect the above 8 SNP sites, and the venous thrombosis risk prediction model established above was used for prediction. The results are as follows.

[0092] Table 4 Clinical validation of the venous thrombosis risk prediction model

[0093] Patient number age gender Weight (kg) History of hypertension History of diabetes Model prediction results Clinical diagnosis (whether venous thrombosis is present) 1 55 female 60 none none No risk No venous thrombosis 2 62 male 75 have none Low risk No venous thrombosis 3 48 female 58 none have No risk No venous thrombosis 4 58 male 72 none none Low risk No venous thrombosis 5 60 female 65 have have Medium risk No venous thrombosis 6 45 male 70 none none No risk No venous thrombosis 7 53 female 55 none none No risk No venous thrombosis 8 65 male 78 have none Medium risk No venous thrombosis 9 50 female 62 none have Low risk No venous thrombosis 10 68 male 75 have have Low risk No venous thrombosis 11 52 female 58 none none No risk No venous thrombosis 12 47 male 68 none none No risk Have venous thrombosis 13 56 female 60 have have High risk Have venous thrombosis 14 63 male 73 none none Low risk No venous thrombosis 15 51 female 56 none none No risk No venous thrombosis 16 49 male 70 none have No risk No venous thrombosis 17 57 female 63 have none High risk Have venous thrombosis 18 61 male 74 have have Low risk No venous thrombosis 19 54 female 59 none none Low risk No venous thrombosis 20 46 male 65 none none No risk No venous thrombosis 21 59 female 61 have have High risk Have venous thrombosis 22 64 male 76 have none Low risk No venous thrombosis 23 50 female 58 none none Low risk No venous thrombosis 24 67 male 77 have have Low risk No venous thrombosis 25 53 female 62 none none Medium risk Have venous thrombosis 26 48 male 69 none none Low risk Have venous thrombosis 27 55 female 64 have have Medium risk No venous thrombosis 28 62 male 71 none none Low risk No venous thrombosis 29 49 female 58 none none Low risk No venous thrombosis 30 66 male 75 have have Medium risk No venous thrombosis 31 54 female 61 none none Low risk Have venous thrombosis 32 47 male 67 none none Low risk No venous thrombosis 33 58 female 63 have have High risk Have venous thrombosis 34 63 male 74 none none Low risk No venous thrombosis 35 52 female 59 none none Low risk No venous thrombosis 36 46 male 64 none none Low risk No venous thrombosis 37 57 female 62 have have High risk Have venous thrombosis 38 61 male 73 have none High risk Have venous thrombosis 39 55 female 60 none none Low risk No venous thrombosis 40 60 male 72 have have Low risk No venous thrombosis 41 53 female 58 none none Medium risk Have venous thrombosis 42 48 male 68 none none Low risk No venous thrombosis 43 56 female 61 have have Low risk No venous thrombosis 44 62 male 74 none none Medium risk Have venous thrombosis 45 51 female 57 none none Low risk No venous thrombosis 46 49 male 69 none none Low risk Have venous thrombosis 47 58 female 62 have have Medium risk No venous thrombosis 48 63 male 75 none none Low risk No venous thrombosis 49 50 female 58 none none Medium risk Have venous thrombosis 50 67 male 76 have have Low risk No venous thrombosis 51 54 female 61 none none Low risk Have venous thrombosis 52 47 male 66 none none Medium risk Have venous thrombosis 53 59 female 61 have have Low risk No venous thrombosis 54 64 male 77 have none Low risk No venous thrombosis 55 53 female 62 none none Medium risk Have venous thrombosis 56 48 male 68 none none Low risk Have venous thrombosis 57 55 female 64 have have Low risk No venous thrombosis 58 62 male 73 none none Low risk No venous thrombosis 59 49 female 58 none none Medium risk Have venous thrombosis 60 66 male 75 have have Low risk No venous thrombosis 61 54 female 61 none none Low risk No venous thrombosis 62 47 male 67 none none Low risk No venous thrombosis 63 58 female 63 have have High risk Have venous thrombosis 64 63 male 74 none none Medium risk Have venous thrombosis 65 52 female 59 none none Low risk No venous thrombosis 66 46 male 64 none none Low risk No venous thrombosis 67 57 female 62 have have High risk Have venous thrombosis 68 61 male 73 have none Medium risk Have venous thrombosis 69 55 female 60 none none Low risk No venous thrombosis 70 60 male 72 have have Low risk No venous thrombosis 71 53 female 58 none none Medium risk Have venous thrombosis 72 48 male 68 none none Medium risk Have venous thrombosis 73 56 female 61 have have Medium risk No venous thrombosis 74 62 male 74 none none Medium risk Have venous thrombosis 75 51 female 57 none none Low risk No venous thrombosis 76 49 male 69 none none Low risk Have venous thrombosis 77 58 female 62 have have High risk Have venous thrombosis 78 63 male 75 none none Low risk No venous thrombosis 79 50 female 58 none none Low risk No venous thrombosis 80 67 male 76 have have High risk Have venous thrombosis 81 54 female 61 none none Medium risk Have venous thrombosis 82 47 male 66 none none Low risk No venous thrombosis 83 59 female 61 have have Medium risk No venous thrombosis 84 64 male 77 have none Low risk Have venous thrombosis 85 53 female 62 none none Medium risk Have venous thrombosis 86 48 male 68 none none Low risk No venous thrombosis 87 55 female 64 have have Low risk No venous thrombosis 88 62 male 73 none none Medium risk Have venous thrombosis 89 49 female 58 none none Low risk Have venous thrombosis 90 66 male 75 have have Low risk No venous thrombosis 91 54 female 61 none none Medium risk No venous thrombosis 92 47 male 67 none none Medium risk Have venous thrombosis 93 58 female 63 have have High risk Have venous thrombosis 94 63 male 74 none none Low risk No venous thrombosis 95 52 female 59 none none Medium risk Have venous thrombosis 96 46 male 64 none none Low risk No venous thrombosis 97 57 female 62 have have High risk Have venous thrombosis 98 61 male 73 have none Low risk No venous thrombosis 99 55 female 60 none none Medium risk Have venous thrombosis 100 60 male 72 have have Low risk No venous thrombosis 101 53 female 58 none none Low risk Have venous thrombosis 102 48 male 68 none none Medium risk Have venous thrombosis 103 56 female 61 have have Low risk No venous thrombosis 104 62 male 74 none none Medium risk Have venous thrombosis 105 51 female 57 none none Medium risk Have venous thrombosis 106 49 male 69 none none High risk Have venous thrombosis 107 58 female 62 have have Medium risk No venous thrombosis

[0094] (2) Calculation of statistical results

[0095] In the prediction model, the sensitivity of high-risk prediction was 29.55%, the specificity was 100%, and the overall accuracy was 71.03%. The sensitivity of medium- and high-risk prediction was 77.27%, the specificity was 85.71%, and the overall accuracy was 82.24%.

[0096] (3) Results analysis

[0097] Through clinical application and verification of these 107 patients, the venous thrombosis risk prediction model showed high sensitivity and specificity, indicating that the model has a certain reliability in identifying patients at risk of venous thrombosis. However, there is still room for further optimization and improvement, such as increasing the amount of data, adjusting model parameters, or improving feature selection methods to improve the performance of the model. At the same time, combined with other clinical indicators and detection methods, the accuracy and reliability of venous thrombosis risk prediction can be further improved.

[0098] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described with reference to the preferred embodiments of the present invention, it should be understood by those skilled in the art that various changes may be made in form and details without departing from the spirit and scope of the present invention as defined in the appended claims.

Claims

1. A probe for detecting SNPs associated with venous thrombosis, characterized in that: After fluorescently labeled polycyclic aromatic bases are modified onto Triplex DNA designed based on target SNPs, they are combined with silicon-based nanomaterials to form probes; the target SNPs include rs6025, rs8176719, rs1799963, rs2066865, rs2036914, rs1799889, rs5918 and rs1801133.

2. The probe for detecting SNPs associated with venous thrombosis according to claim 1, characterized in that: The polycyclic aromatic base is obtained by introducing hydroxyl, amino or methyl groups into the base positions of naphthyl polycyclic aromatic bases, anthracenyl polycyclic aromatic bases, phenanthryl polycyclic aromatic bases, pyrene polycyclic aromatic bases, fluoranthene polycyclic aromatic bases or benzanthracenyl polycyclic aromatic bases.

3. The probe for detecting venous thrombosis-related SNP according to claim 2, characterized in that: The polycyclic aromatic base has a three-membered ring structure.

4. The probe for detecting SNPs associated with venous thrombosis according to claim 3, characterized in that: The number of polycyclic aromatic bases modified with fluorescently labeled polycyclic aromatic bases on the Triplex DNA designed by the target SNP is 1 or 2; when the number of the polycyclic aromatic bases is 1, the length of the Triplex DNA is 25bp; when the number of the polycyclic aromatic bases is 2, the length of the Triplex DNA is 15bp.

5. The probe for detecting SNPs associated with venous thrombosis according to claim 4, characterized in that: The polycyclic aromatic base is located 2-3 bp from the 5' end of the target SNP, and the interval between the multiple polycyclic aromatic bases is greater than or equal to 3 bp.

6. The probe for detecting SNPs associated with venous thrombosis according to claim 5, characterized in that: The nucleotide sequence of the TriplexDNA is shown as SEQ ID NO.1-SEQ ID NO.

8.

7. A risk prediction system for venous thrombosis, characterized in that: It includes data acquisition module, model building module, SNP detection module and prediction module; The data acquisition module is used to collect the target SNP data and clinical characteristic data of patients with venous thrombosis as claimed in claim 1 as a training set; The model building module is used to train the venous thrombosis risk prediction model based on the obtained training set using a random forest algorithm to learn the relationship between the target SNP test results and clinical feature data and the venous thrombosis risk; The SNP detection module is used to detect whether the target SNP exists in the subject based on the probe for venous thrombosis-related SNP detection according to any one of claims 1 to 5, using the wavelength shift data of the fluorescent marker, to obtain the target SNP detection result of the subject; The prediction module is used to input the obtained target SNP detection results and clinical characteristic data of the subject into the venous thrombosis risk prediction model to obtain a venous thrombosis risk score.

8. The risk prediction system for venous thrombosis according to claim 7, characterized in that: The clinical characteristic data include age, gender, weight, history of hypertension and history of diabetes.

9. The risk prediction system for venous thrombosis according to claim 7, characterized in that: The lowest concentration of the target SNP in the test sample of the subject is 0.5 pM.

10. Use of the probe for detecting venous thrombosis-related SNPs according to any one of claims 1 to 6 in preparing a venous thrombosis detection reagent.

Images