TERT promoter expression state prediction method and device, medium and product
By constructing a prediction model of imagingomics and clinical data, the problem of predicting the expression status of TERT promoter in brain glioma is solved, and the accurate prediction of the expression status of TERT promoter is achieved, supporting molecular typing and individualized treatment decisions of glioma.
Patent Information
- Application Number
- CN202510208073.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-02-25
- Publication Date
- 2025-06-13
AI Technical Summary
The prior art is difficult to effectively predict the expression status of TERT promoter in brain gliomas, especially in isocitrate dehydrogenase wild-type low-grade gliomas (IDHwt LGA).
By combining imagingomics and clinical data, two predictive models were constructed using support vector machines, random forests or linear regression models, and target characteristics were extracted from imagingomics and clinical features, respectively, and the TERT promoter expression status was predicted.
Accurate prediction of the expression status of TERT promoter of brain glioma is achieved, and non-invasive detection tools are provided to support molecular typing and prognostic evaluation of gliomas, help individualized treatment decisions, and improve patients' clinical prognosis and quality of life.
Smart Images
Figure CN120148775A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the technical field of radiomics prediction, and particularly to a method, device, medium and product for predicting the expression status of the TERT promoter. Background Art
[0002] Glioblastoma is one of the most common malignant brain tumors, and its clinical characteristics and prognosis are closely related to the molecular characteristics of the tumor. Approximately 30% of glioblastomas are low-grade gliomas. Although low-grade gliomas usually have a good prognosis, almost all low-grade gliomas will eventually progress to high-grade gliomas. This transformation process is accompanied by obvious clinical deterioration and a high mortality rate. Therefore, it is crucial to identify early whether low-grade gliomas have the potential for malignant transformation to improve patient survival and optimize treatment plans.
[0003] Telomerase reverse transcriptase (TERT) promoter mutations have been proven to be closely related to the malignant process of glioblastoma, especially in low-grade gliomas. These mutations usually promote the proliferation of tumor cells by enhancing the expression of TERT, thereby accelerating the malignant progression of the tumor. TERT promoter mutations are clinically used as an important marker for judging glioma grade and proliferation activity. Although TERT promoter mutations can provide important molecular basis for glioma grading and prognosis, the role of current imaging biomarkers in predicting these molecular characteristics is still understudied.
[0004] For the prediction of TERT promoter mutations (TERTp mutations) in glioblastoma, especially in isocitrate dehydrogenase wild-type low-grade gliomas (IDHwt LGA), the application of radiomics combined with clinical data has become a non-invasive detection tool with great potential. Radiomics can effectively reflect the biological behavior of tumors, including tumor growth, invasiveness and molecular characteristics, by extracting a large number of high-dimensional imaging features from multi-modal imaging data. The radiomics-based method for predicting TERT promoter mutations in glioblastoma can extract rich quantitative features from multi-modal magnetic resonance images and analyze them through machine learning algorithms to predict whether a patient has TERTp mutations. This non-invasive detection method can not only provide important support for the molecular typing and prognosis assessment of gliomas, but also provide a scientific basis for individualized treatment decisions, and ultimately is expected to improve the clinical prognosis and quality of life of patients. Although recent studies have shown the potential of radiomics in predicting the molecular characteristics of nervous system tumors, such as meningiomas, pineal region tumors and gliomas with isocitrate dehydrogenase mutations, the role of radiomics in predicting TERTp mutations in IDHwt LGA has not been fully explored. Summary of the Invention
[0005] The objective of this application is to provide a method, device, medium and product for predicting the expression status of the TERT promoter, so as to solve the problem of being unable to predict the expression status of the TERT promoter in gliomas.
[0006] To achieve the above objective, this application provides the following solutions:
[0007] In the first aspect, this application provides a method for predicting the expression status of the TERT promoter, including:
[0008] Determine the target radiomics features and target clinical features of the target patient; the target patient is a patient with resected low-grade glioma to be predicted;
[0009] Input the target radiomics features of the target patient into the first expression status prediction model to obtain the first prediction value of the expression status of the TERT promoter of the target patient; the first expression status prediction model is constructed using a support vector machine, random forest or linear regression model;
[0010] Input the target clinical features of the target patient into the second expression status prediction model to obtain the second prediction value of the expression status of the TERT promoter of the target patient; the second expression status prediction model is constructed using a support vector machine, random forest or linear regression model;
[0011] Based on the first prediction value and the second prediction value of the expression status of the TERT promoter of the target patient, determine the final prediction value of the expression status of the TERT promoter of the target patient; the expression status of the TERT promoter is mutated or not mutated.
[0012] Optionally, determining the target radiomics features and target clinical features of the target patient includes:
[0013] Obtain the original radiomics features and original clinical features of the target patient;
[0014] Use a screening algorithm to screen the original radiomics features and original clinical features of the target patient respectively to obtain the target radiomics features and target clinical features of the target patient.
[0015] Optionally, the original radiomics features include: T1, T2, SWI, FLAIR, DWI, T2WI, CE-T1WI, T1 enhancement, FA map and ADC map.
[0016] Optionally, the original clinical features include: WHO grade, age, gender, midline shift, laterality, peritumoral edema, enhancing margin, enhancement type, enhancement degree and MGMT methylation status.
[0017] Optionally, the screening algorithm is a multi-fold loop algorithm, lasso regression algorithm, elastic regression algorithm, T-test, network model, or weight analysis method.
[0018] Optionally, the process of determining the first expression status prediction model includes:
[0019] Obtain a first training set; the first training set includes: the original radiomics features of multiple sample patients and the actual values of the TERT promoter expression status; the sample patients are patients who have undergone resection of low-grade gliomas;
[0020] Use the screening algorithm to screen the original radiomics features of each sample patient to obtain the target radiomics features of each sample patient;
[0021] Determine a support vector machine, random forest, or linear regression model as the first initial network;
[0022] Use the first training set to train the first initial network to obtain the first expression status prediction model.
[0023] Optionally, the process of determining the second expression status prediction model includes:
[0024] Obtain a second training set; the second training set includes: the original clinical features of multiple sample patients and the actual values of the TERT promoter expression status; the sample patients are patients who have undergone resection of low-grade gliomas;
[0025] Use the screening algorithm to screen the original clinical features of each sample patient to obtain the target clinical features of each sample patient;
[0026] Determine a support vector machine, random forest, or linear regression model as the second initial network;
[0027] Use the second training set to train the second initial network to obtain the second expression status prediction model.
[0028] In a second aspect, the present application provides a computer device, including: a memory, a processor, and a computer program stored on the memory and executable on the processor, where the processor executes the computer program to implement the TERT promoter expression status prediction method described in any one of the above.
[0029] In a third aspect, the present application provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, it implements the TERT promoter expression status prediction method described in any one of the above.
[0030] Fourthly, the present application provides a computer program product, including a computer program which, when executed by a processor, implements the TERT promoter expression status prediction method described in any one of the above.
[0031] According to the specific embodiments provided by the present application, the following technical effects are disclosed:
[0032] The present application discloses a method, device, medium and product for predicting the expression status of the TERT promoter. First, determine the target radiomics features and target clinical features of the target patient; the target patient is a patient with resected low-grade glioma to be predicted; secondly, input the target radiomics features of the target patient into the first expression status prediction model to obtain the first predicted value of the TERT promoter expression status of the target patient; the first expression status prediction model is constructed using a support vector machine, random forest or linear regression model; then, input the target clinical features of the target patient into the second expression status prediction model to obtain the second predicted value of the TERT promoter expression status of the target patient; the second expression status prediction model is constructed using a support vector machine, random forest or linear regression model; finally, based on the first predicted value and the second predicted value of the TERT promoter expression status of the target patient, determine the final predicted value of the TERT promoter expression status of the target patient; the TERT promoter expression status is mutated or not mutated. The present application combines the radiomics features and clinical features of the patient to realize the prediction of the TERT promoter expression status of glioma. Description of the Drawings
[0033] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the following will briefly introduce the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.
[0034] Figure 1 Schematic flowchart of the method for predicting the expression status of the TERT promoter provided by an embodiment of the present application;
[0035] Figure 2 Schematic structural diagram of a computer device provided by an embodiment of the present application. Detailed Embodiments
[0036] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the present application without creative efforts belong to the scope of protection of the present application.
[0037] The purpose of the present application is to provide a method, device, medium and product for predicting the expression status of the TERT promoter, aiming to realize the prediction of the expression status of the TERT promoter in glioma.
[0038] To make the above objects, features and advantages of the present application more obvious and understandable, the present application will be further described in detail below in conjunction with the accompanying drawings and specific embodiments.
[0039] In an exemplary embodiment, as Figure 1 shown, a method for predicting the expression status of the TERT promoter is provided, including:
[0040] Step 1: Determine the target radiomics features and target clinical features of the target patient; the target patient is a patient with resected low-grade glioma to be predicted.
[0041] As an optional implementation manner, Step 1 includes:
[0042] Step 11: Obtain the original radiomics features and original clinical features of the target patient.
[0043] As an optional implementation manner, the original radiomics features include: T1, T2, Susceptibility Weighted Imaging (SWI), FLAIR, Diffusion-Weighted Imaging (DWI), T2-Weighted Imaging (T2WI), Contrast-Enhanced T1WI (CE-T1WI), T1 enhancement, Fractional Anisotropy (FA) map, and Apparent Diffusion Coefficient (ADC) map.
[0044] Among them, the FA map is an imaging index used in MRI, usually used in combination with diffusion tensor imaging to reflect the diffusion characteristics of water molecules in anisotropic structures.
[0045] As an optional implementation manner, the original clinical features include: WHO grade, age, gender, midline shift, laterality, peritumoral edema, enhancing margin, enhancement type, enhancement degree, and MGMT methylation status.
[0046] Step 12: Use a screening algorithm to screen the original radiomics features and original clinical features of the target patient respectively, and obtain the target radiomics features and target clinical features of the target patient.
[0047] As an alternative implementation, the screening algorithm is a multi-fold loop algorithm, lasso regression algorithm, elastic regression algorithm, T-test, network model or weight analysis method.
[0048] Step 2: Input the target radiomics features of the target patient into the first expression status prediction model to obtain the first predicted value of the TERT promoter expression status of the target patient.
[0049] Among them, the first expression status prediction model is constructed using a support vector machine, random forest or linear regression model.
[0050] As an alternative implementation, in Step 2, the determination process of the first expression status prediction model includes:
[0051] Step 21: Obtain a first training set; the first training set includes: the original radiomics features of multiple sample patients and the actual values of the TERT promoter expression status; the sample patients are patients who have undergone resection of low-grade gliomas.
[0052] Specifically, the inclusion criteria for sample patients specifically include: (I) patients diagnosed with IDHwt LGA by histopathology; (II) patients with known TERTp mutation status and IDH wild-type status; (III) patients after initial tumor resection.
[0053] The exclusion criteria for sample patients specifically include: (a) patients lacking clinical information or with incomplete magnetic resonance imaging (MRI) sequences; (b) patients under 18 years old; (c) patients with image processing errors; (d) patients with unknown TERTp mutation status.
[0054] Step 22: Use a screening algorithm to screen the original radiomics features of each sample patient to obtain the target radiomics features of each sample patient.
[0055] Specifically, during a certain prediction, T2W and CE-T1WI in the target radiomics features include 1,304 radiomics features. The radiomics features include 14 first-order statistics, 126 shape features, 688 wavelet features, and 476 texture features. Among them, the first-order statistics use common and basic indicators to describe the distribution of voxel intensities within the image region defined by the mask; the shape features mainly describe the three-dimensional size and shape of the tumor region; the wavelet features can effectively separate the texture features from the high-frequency and low-frequency components of the original image; the texture features describe the patterns or spatial distributions of voxel intensities and have four types: gray-level co-occurrence matrix, gray-level run-length matrix, gray-level size zone matrix, and gray-level dependence matrix.
[0056] Specifically, T2WI and CE-T1WI are axial T2WI and axial CE-T1WI. Before the examination, all patients signed an informed consent form and removed all metal objects. Also, the examination was performed at the spinal position using a 3.0-T scanner (Siemens Magnetom Skyra). For T2WI and CE-T1WI, the repetition time was 4,900 ms and 1,770 ms; the echo time was 117.12 ms and 9.4 ms; the acquisition matrix was 320×288 and 256×198 respectively; the flip angle was 90° and 150° respectively. In addition, the field of view, slice thickness, and slice gap were all 100×100, 5 mm, and 6 mm. In addition, after the patient was injected with a contrast agent (gadolinium-DTPA, Magnevist, 0.1 mmol / kg), CE-T1WI was performed with the parameters of T1WI.
[0057] Step 23: Determine the support vector machine, random forest, or linear regression model as the first initial network.
[0058] Step 24: Use the first training set to train the first initial network to obtain the first expression state prediction model.
[0059] Specifically, Step 24 includes:
[0060] Step 241: Use the normalization formula to normalize each target radiomics feature in the first training set. The normalization formula is:
[0061]
[0062] where X' n is the nth normalized target radiomics feature X; X n is the nth target radiomics feature X; X min is the minimum value of all target radiomics features X; X max is the maximum value of all target radiomics features X.
[0063] Step 242: Using the normalized target radiomics features of each sample patient as input and the actual value of the corresponding TERT promoter expression status as output, train the first initial network to obtain the first expression status prediction model.
[0064] Specifically, the support vector machine creates a high-dimensional hyperplane, where different categories are divided into 17 categories. The larger the functional margin, the better the hyperplane is separated, which means the lower the generalization error. Regression and classification tasks can be achieved through the support vector machine. The parameters determined by the GridSearch method are the best-performing parameters in the model. In addition, violin plots were used to compare the differences between TERTp mutant and TERTp wild-type patients in IDHwt LGA.
[0065] The support vector machine improves the generalization ability of the learning machine by seeking the minimum of the structural risk, realizes the minimization of the empirical risk and the confidence range, so as to achieve the purpose of obtaining good statistical laws even when the statistical sample size is small. The support vector machine greatly improves the generalization performance of the prediction model by selecting the hyperplane with the largest margin, and at the same time, using kernel functions such as the radial basis function can establish a non-linear model. Through the non-linear modeling of the support vector machine, an effective first expression status prediction model is finally established. The linear regression model is used to calculate the role (weight) of each feature in radiomics in the prediction, and the features with large weights are extracted through screening for establishing the first expression status prediction model.
[0066] Step 3: Input the target clinical features of the target patient into the second expression status prediction model to obtain the second predicted value of the TERT promoter expression status of the target patient.
[0067] Among them, the second expression status prediction model is constructed using a support vector machine, random forest or linear regression model.
[0068] As an optional implementation manner, in Step 3, the determination process of the second expression status prediction model includes:
[0069] Step 31: Obtain a second training set; the second training set includes: the original clinical features of multiple sample patients and the actual values of the TERT promoter expression status; the sample patients are patients who have undergone resection of low-grade gliomas.
[0070] Step 32: Use a screening algorithm to screen the original clinical features of each sample patient to obtain the target clinical features of each sample patient.
[0071] Step 33: Determine the support vector machine, random forest or linear regression model as the second initial network.
[0072] Specifically, the formula of the linear regression model is:
[0073] riskscore = Σcoefficient n ×β n 。
[0074] Wherein, riskscore is the predicted value (the first predicted value or the second predicted value) of the TERT promoter expression status obtained after calculation by the linear regression model; coefficient n is the nth target radiomics feature or target clinical feature; β n is the coefficient corresponding to the nth target radiomics feature or target clinical feature.
[0075] Step 34: Train the second initial network using the second training set to obtain a second expression status prediction model.
[0076] Step 4: Based on the first predicted value and the second predicted value of the TERT promoter expression status of the target patient, determine the final predicted value of the TERT promoter expression status of the target patient; the TERT promoter expression status is mutated or not mutated.
[0077] Specifically, Step 4 includes:
[0078] Step 41: When both the first predicted value and the second predicted value of the TERT promoter expression status of the target patient are mutated, determine the final predicted value of the TERT promoter expression status of the target patient as mutated.
[0079] Step 42: When both the first predicted value and the second predicted value of the TERT promoter expression status of the target patient are not mutated, determine the final predicted value of the TERT promoter expression status of the target patient as not mutated.
[0080] Furthermore, to establish a more accurate and comprehensive tumor discrimination model, the present application also constructed a combined radiomics nomogram including a first expression status prediction model and a second expression status prediction model by using multivariate logistic regression analysis. The nomogram was plotted using the "rms" software package (version 6.3.0) to illustrate the structure and parameters of the first expression status prediction model. The ROC curve was plotted using the "pROC" software package on the training set and the validation set, and the area under the ROC curve (AUC), accuracy (ACC), sensitivity, positive predictive value (PPV), and negative predictive value (NPV) were obtained from the ROC curve; to evaluate the similarity between the observed results and the predicted diagnostic results of the first expression status prediction model and the combined radiomics nomogram, a calibration curve and Hosmer-Lemeshow test were performed; based on the DCA method of the "rmda" software package (version 1.6), the net benefit of the first expression status prediction model, the second expression status prediction model, and the combined radiomics nomogram at different threshold probabilities was quantified. Finally, the De long test was performed to compare the statistical significance of the AUC values among the clinical model, the first expression status prediction model, and the combined radiomics nomogram. Finally, the final predicted value was determined from the combined radiomics nomogram.
[0081] In an exemplary embodiment, a computer device is provided, including: a memory, a processor, and a computer program stored on the memory and executable on the processor, and the processor executes the computer program to implement the TERT promoter expression status prediction method.
[0082] In an exemplary embodiment, a computer-readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the TERT promoter expression status prediction method is implemented.
[0083] In an exemplary embodiment, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, the TERT promoter expression status prediction method is implemented.
[0084] In an exemplary embodiment, a computer device is provided, and the computer device can be a server or a terminal, and its internal structure diagram can be as Figure 2As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, abbreviated as I / O), and a communication interface. Among them, the processor, the memory, and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and external devices. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, it realizes a method for predicting the expression state of the TERT promoter.
[0085] Those skilled in the art can understand that Figure 2 the structure shown in is only a block diagram of some structures related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.
[0086] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, database, or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memories. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc.
[0087] The databases involved in the embodiments provided in the present application can include at least one of relational databases and non-relational databases. Non-relational databases can include distributed databases based on blockchain, etc., without limitation. The processors involved in the embodiments provided in the present application can be general-purpose processors, central processors, graphics processors, digital signal processors, programmable logic devices, data processing logics based on quantum computing, etc., without limitation.
[0088] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in the present application are all information and data authorized by the user or fully authorized by all parties, and the collection, use, and processing of relevant data need to comply with relevant regulations.
[0089] The technical features of the above embodiments can be combined arbitrarily. For the sake of concise description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered to be within the scope described in this specification.
[0090] In this text, specific examples are used to illustrate the principle and implementation manner of the present application. The description of the above embodiments is only used to help understand the method and its core idea of the present application; at the same time, for those of ordinary skill in the art, according to the idea of the present application, there will be changes in the specific implementation manner and application scope. In summary, the content of this specification should not be construed as a limitation to the present application.
Claims
1. A method for predicting the expression status of a TERT promoter, characterized in that: The TERT promoter expression status prediction method comprises: Determining target radiomics characterization and target clinical features of a target patient; the target patient is a patient who is predicted to undergo resection of a low-grade brain glioma; Inputting the target radiomics representation of the target patient into a first expression state prediction model to obtain a first prediction value of the TERT promoter expression state of the target patient; the first expression state prediction model is constructed using a support vector machine, a random forest or a linear regression model; Inputting the target clinical features of the target patient into a second expression status prediction model to obtain a second predicted value of the TERT promoter expression status of the target patient; the second expression status prediction model is constructed using a support vector machine, a random forest or a linear regression model; Based on the first prediction value and the second prediction value of the TERT promoter expression state of the target patient, a final prediction value of the TERT promoter expression state of the target patient is determined; the TERT promoter expression state is mutation or non-mutation.
2. The method for predicting TERT promoter expression status according to claim 1, characterized in that: Identify target radiomics characterization and target clinical features of target patients, including: Obtain original radiomics characterization and original clinical characteristics of target patients; The original radiomics representation and original clinical characteristics of the target patients are screened respectively by using the screening algorithm to obtain the target radiomics representation and target clinical characteristics of the target patients.
3. The method for predicting the expression status of TERT promoter according to claim 2, characterized in that: The original imaging omics characterization includes: T1, T2, SWI, FLAIR, DWI, T2WI, CE-T1WI, T1 enhancement, FA map and ADC map.
4. The method for predicting the expression status of TERT promoter according to claim 3, characterized in that: The original clinical characteristics included: WHO grade, age, gender, midline shift, laterality, peritumoral edema, enhancement margin, enhancement type, enhancement degree and MGMT methylation status.
5. The method for predicting the expression status of TERT promoter according to claim 4, characterized in that: The screening algorithm is a multi-fold cycle algorithm, a lasso regression algorithm, an elastic regression algorithm, a T test, a network model or a weight analysis method.
6. The method for predicting the expression status of TERT promoter according to claim 5, characterized in that: The process of determining the first expression status prediction model includes: Obtaining a first training set; the first training set includes: original imaging genomics representations of multiple sample patients and actual values of TERT promoter expression status; the sample patients are patients who have undergone resection of low-grade brain gliomas; The original imaging omics representation of each sample patient is screened by using a screening algorithm to obtain the target imaging omics representation of each sample patient; Determine a support vector machine, a random forest, or a linear regression model as a first initial network; The first initial network is trained using the first training set to obtain the first expression state prediction model.
7. The method for predicting the expression status of TERT promoter according to claim 5, characterized in that: The process of determining the second expression status prediction model includes: Obtaining a second training set; the second training set includes: original clinical characteristics of multiple sample patients and actual values of TERT promoter expression status; the sample patients are patients who have undergone resection of low-grade brain gliomas; The original clinical characteristics of each sample patient are screened using a screening algorithm to obtain the target clinical characteristics of each sample patient; Determine a support vector machine, a random forest, or a linear regression model as a second initial network; The second initial network is trained using the second training set to obtain the second expression state prediction model.
8. A computer device comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the method for predicting the expression status of the TERT promoter according to any one of claims 1 to 7.
9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for predicting the expression status of the TERT promoter according to any one of claims 1 to 7 is implemented.
10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the method for predicting the expression status of the TERT promoter according to any one of claims 1 to 7 is implemented.