Method and device for predicting LUAD prognosis risk, medium and program product
Through the whole gene expression profile data and machine learning algorithm, a prognostic risk model for LUAD tumors was constructed, and the molecular subtype classification of stage IB patients was refined, which solved the problem of lack of refinement of molecular subtype classification in the existing technology, and achieved the Effective assessment of prognostic risks in LUAD patients and optimization of treatment strategies.
Patent Information
- Application Number
- CN202411264594.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-10
- Publication Date
- 2024-12-03
AI Technical Summary
The existing risk assessment of prognostic risk of LUAD tumors depends on macropathological characteristics, and the molecular subtype classification lacks refinement, making it difficult to provide effective clinical guidance.
Cluster analysis of patients with stage IB LUAD through whole gene expression profile data, using machine learning algorithms to screen key genes, and construct prognostic risk models, including three subtypes: IB-L, IB-M and IB-H, using CENPE and FSCN1 , SLC2A1 and other genes are subject to risk scores to predict and classify patients' prognostic risks.
The constructed prognostic risk model can effectively differentiate stage IB LUAD patients into high-risk, medium-risk and low-risk subtypes, providing new clinical guidance, improving the accuracy of prognostic evaluation and targeting treatment strategies.
Smart Images

Figure CN119069005A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of intelligent medical care, and more specifically, to a method, device, medium and program product for predicting the prognosis risk of LUAD. Background Art
[0002] LUAD stands for lung adenocarcinoma, a specific type of lung cancer. Lung adenocarcinoma is one of the most common types of lung cancer and usually originates from the bronchial cells in the lungs, especially the cells that secrete mucus.
[0003] Currently, the prognostic characteristics of the disease and whether adjuvant therapy should be used after surgery are still controversial. Tumor classification, that is, screening for different risk subtypes, is the key to precision medicine. Some stage IB LUAD tumors (4cm-5cm) in the 7th edition TNM staging system of the International Association for the Study of Lung Cancer (IASLC) were upgraded to stage IIA in the 8th edition staging system. According to the 2021 National Comprehensive Cancer Network (NCCN) guidelines, postoperative adjuvant chemotherapy can be considered for stage IB patients with high-risk factors, such as poorly differentiated tumors (including pulmonary neuroendocrine tumors, excluding well-differentiated neuroendocrine tumors), vascular invasion, wedge resection, and visceral pleural involvement.
[0004] However, the existing high-risk factors are more based on macroscopic pathological features, with relatively few distinguishing dimensions (pathological classification mainly depends on microscopic cell and tissue morphological characteristics, such as different cell types, tissue structure, vascular pleural invasion, etc., and the available input parameters are limited). Molecular classification helps to understand tumor heterogeneity and improve clinical treatment. The molecular subtypes of LUAD are defined by a combination of transcriptional subtypes and histopathological, anatomical and mutation classifications, which also have important clinical significance. The upcoming 9th edition of lung adenocarcinoma has collected molecular information to evaluate its prognostic value and the feasibility of incorporating this information into the TNM classification system. However, the molecular classification of stage IB LUAD is still lacking. Therefore, refining the molecular classification has obvious clinical guidance significance for the classification of stage IB LUAD patients. Summary of the invention
[0005] The present invention aims to solve at least one of the technical problems existing in the prior art. To this end, the present invention provides a method for predicting the prognosis risk of LUAD; the method of the present invention uses the whole gene expression spectrum data to classify stage IB LUAD patients for the first time, which has obvious clinical guidance significance.
[0006] The first aspect of the present application discloses a method for constructing a prognostic risk model for stage IB LUAD, the method comprising:
[0007] S1, obtain the gene expression data of the IB stage training set samples and the corresponding survival status of the samples;
[0008] S2, performing cluster analysis on the IB stage training set samples based on the survival status to obtain different subtypes including the first IB-L and the first IB-H and the gene expression status characterizing each subtype;
[0009] S3, screening the genes in the first IB-H subtype and the first IB-L subtype by a machine learning algorithm to obtain IB-H and IB-L key genes, respectively;
[0010] S4, taking the intersection of the IB-H key gene and the IB-L key gene to obtain the predicted gene; constructing the different subtypes including the second IB-L, the second IB-M and the second IB-H and the prognostic risk model characterizing the gene expression of each group of subtypes based on the predicted gene.
[0011] In some embodiments, the different subtypes in S2 and the gene expression conditions representing each group of subtypes further include: a first IB-M subtype and the gene expression conditions representing the subtype;
[0012] Optionally, the cluster analysis in S2 includes: clustering the IB stage training set samples according to the similarity of the gene expression data to obtain clustering results including any one or more of IB1, IB2, IB3 and IB4; subdividing any one or more of the clustering results of IB1, IB2, IB3 and IB4 into any one or more of the clustering results of IB-L, IB-M and IB-H according to the survival status of the IB stage training set samples;
[0013] Optionally, the machine learning algorithm includes any one or more of the following: rf, gbm, XGBoost, LASSO;
[0014] Optionally, the method of clustering the samples of the IB stage training set according to the similarity of the gene expression data includes: consistency clustering (CDF);
[0015] Optionally, the prognostic risk model constructed based on the predicted genes is established based on logistic regression.
[0016] In some embodiments, the predicted genes include at least two of CENPE (centromere protein E), FSCN1 (fascin actin-bundling protein 1), and SLC2A1 (solute carrier family 2 member 1);
[0017] Optionally, the risk score of the predicted gene corresponding to the second IB-L is in the L interval, the risk score of the predicted gene corresponding to the second IB-M is in the M interval, and the risk score of the predicted gene corresponding to the second IB-H is in the H interval; that is, the risk scores of the predicted genes corresponding to the second IB-L, the second IB-M and the second IB-H are from low to high;
[0018] Optionally, the L interval, M interval and H interval are obtained by training the prediction genes in the training set samples and the different subtypes of the samples;
[0019] Optionally, risk score = CENPE*a+FSCN1*b+SLC2A1*c, where a is the coefficient of CENPE, b is the coefficient of FSCN1, and c is the coefficient of SLC2A1.
[0020] The second aspect of the present application discloses a method for predicting the prognosis risk of LUAD, the method comprising:
[0021] Obtaining gene expression data of the sample to be tested;
[0022] Inputting the gene expression data of the sample to be tested into the prognostic risk model disclosed in the first aspect of the present application to obtain any one of the classification results of the second IB-L, the second IB-M and the second IB-H;
[0023] Optionally, the gene expression data includes any two or three of the following genes: CENPE, FSCN1, SLC2A1.
[0024] The third aspect of the present application is a method for predicting the prognosis risk of LUAD, the method comprising:
[0025] Acquiring expression data of predicted genes in the sample to be tested; the predicted genes include any two or three of CENPE (centromere protein E), FSCN1 (fascin actin-bundling protein 1), and SLC2A1 (solute carrier family 2 member 1);
[0026] Calculating the expression levels of the at least two prediction genes, and classifying the samples to be tested according to the expression levels to obtain any one of the classification results of the second IB-L, the second IB-M, and the second IB-H;
[0027] Optionally, the predicted gene expressions are all negative and are defined as the second IB-L, the predicted gene expressions are all positive and are defined as the second IB-H, and non-second IB-L and non-IB-H are defined as the second IB-M.
[0028] In some embodiments, the survival period of the sample to be tested is predicted based on the classification result; based on the result of the second IB-L, a result indicating that the survival period of the sample to be tested is long is output; based on the result of the second IB-H, a result indicating that the survival period of the sample to be tested is short is output; based on the result of the second IB-M, a result indicating that the survival period of the sample to be tested is between the second IB-L and the second IB-H is output;
[0029] Optionally, the method also includes: based on the result of the second IB-L, outputting a result that the survival period of the sample to be tested is similar to stage IA; based on the result of the second IB-H, outputting a result that the survival period of the sample to be tested is similar to stage II; based on the result of the second IB-M, outputting a result that the survival period of the sample to be tested is between stage IA and stage II.
[0030] In some embodiments, the method further includes: based on the result of the second IB-L, outputting the result that the sample to be tested can refer to the result of the stage IA treatment strategy; based on the result of the second IB-H, outputting the result that the sample to be tested can refer to the result of the stage II treatment strategy; based on the result of the second IB-M, outputting the result that the sample to be tested needs to comprehensively consider the treatment mode.
[0031] The fourth aspect of the present application discloses a computer device, which includes: a memory and a processor; the memory is used to store a computer program; and the processor executes the computer program to implement the steps of the above method.
[0032] A fifth aspect of the present application discloses a computer-readable storage medium having a computer program stored thereon, which implements the steps of the above method when the computer program is executed by a processor.
[0033] A sixth aspect of the present application discloses a computer program product, including a computer program, which implements the steps of the above method when executed by a processor.
[0034] This application has the following beneficial effects:
[0035] 1. The present application innovatively discloses a method for constructing a prognostic risk model for stage IB LUAD. According to the differences in gene expression profiles and prognostic characteristics, a prognostic risk model is constructed to divide stage IB LUAD patients into stage IB high-risk (BH), intermediate-risk (IBM) and low-risk (IB-L) subtypes. At the same time, during the model construction process, three independent prognostic biomarkers, CENPE, FSCN1, and SLC2A1, were also identified. This model can be used to effectively predict the prognosis of LUAD patients, provide new clues and perspectives for the molecular subtype classification of stage IB LUAD, and has potential clinical significance for the treatment and management strategies of stage IB LUAD patients.
[0036] 2. This application innovatively classifies stage IB LUAD patients based on the above classification model, and uses the analysis of the patients' gene expression data or predictive genes to evaluate the patients' clinical prognosis. This method deeply explores the life laws hidden behind the biological data, and conducts in-depth analysis from multiple dimensions such as the genetic information and pathway information of the biological population, greatly improving the accuracy and depth of data analysis. BRIEF DESCRIPTION OF THE DRAWINGS
[0037] 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.
[0038] Figure 1 is a schematic diagram of a method flow chart provided by the first aspect of an embodiment of the present invention;
[0039] Figure 2 is a schematic diagram of a method flow chart provided by the second aspect of an embodiment of the present invention;
[0040] Figure 3 is a schematic diagram of a method flow chart provided by the third aspect of an embodiment of the present invention;
[0041] Figure 4 is a schematic diagram of a computer device provided by an embodiment of the present invention;
[0042] Figure 5 is a schematic diagram of the architecture of an exemplary computing device provided by an embodiment of the present invention;
[0043] Figure 6 is a schematic diagram of a storage medium provided by an embodiment of the present invention;
[0044] Figure 7 It is a schematic diagram of the design idea provided by an embodiment of the present invention;
[0045] Figure 8 is a schematic diagram of the molecular typing of stage IB LUAD provided in an embodiment of the present invention, wherein: Figure 8 A shows the prognostic genes associated with stage IB LUAD by Cox regression analysis. Figure 8 B shows that TCGA-LUAD patients in stage IB are consistently divided into four types. Figure 8 C shows the classification effect of UMAP (left) and t-SNE (right) for IB1 / 2 / 3 / 4 patients. Figure 8 D shows the OS evaluation of patients in stage IB1 / 2 / 3 / 4;
[0046] Fig. 9: is a schematic diagram of the clinical significance evaluation of the molecular classification of stage IB LUAD patients provided in an embodiment of the present invention, including the OS evaluation between stage IB1 / 2 / 3 / 4 patients and stage IA patients (A) and stage II patients (B), and the OS evaluation between stage IB-L, IB-M and IB-H patients and stage IA patients (C) and stage II patients (D);
[0047] Fig.10 Schematic diagram of the molecular typing discriminant model of stage IB LUAD provided in an embodiment of the present invention, wherein the confusion matrix of the discriminant model in dataset 1 (A: rf, D: gbm and G: xgbTree). OS evaluation between stage IB-L, IB-M and IB-H patients and stage IA patients (B: rf, E: gbm and H: xgbTree) and stage II patients (C: rf, F: gbm and I: xgbTree) in dataset 2;
[0048] Fig.11 Schematic diagram of establishing the risk score of molecular typing of stage IB LUAD provided in an embodiment of the present invention, wherein the importance ranking of discriminating genes predicted by rf (A), gbm (B) and XGBoost (C) (top 30, IB-H and IB-L); (D) LASSO regression filtering discriminating genes (IB-H and IB-L); E) intersection of key discriminant genes (IB-H and IB-L) identified by the above four machine learning algorithms; OS evaluation between IB-H, IB-M and IB-L patients (classified by risk score) and stage IA patients (left) and stage II patients (right) in dataset 1; (G) IB-H, IB-M and IB-L patients (classified by risk score) and stage IA patients (left) and stage II patients (right) in dataset 2;
[0049] Fig.12 Schematic diagram of key genes for molecular classification of stage IB LUAD provided by an embodiment of the present invention, wherein the expression of CENPE (left), FSCN1 (middle) and SLC2A1 (right) in cancer tissues is compared with that in normal tissues (A: Dataset 1; B: Dataset 2); the prognostic significance of the three key genes in Dataset 1 (left) and Dataset 2 (right) (C: CENPE; D: FSCN1 and E: SLC2A1); in Dataset 1, OS evaluation between IB-H, IB-M and IB-L patients (by joint division of the three genes) and stage IA patients (F) and stage II patients (G); in Dataset 2, OS evaluation between IB-H, IB-M and IB-L patients (by joint division of the three genes) and stage IA patients (H) and stage II patients (I). DETAILED DESCRIPTION
[0050] 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.
[0051] In some of the processes described in the specification and claims of the present invention and the above-mentioned figures, multiple operations that appear 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 article or executed in parallel. The serial numbers of the operations, such as 101, 102, 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. It should be noted that the descriptions of "first", "second", etc. in this article are used to distinguish different messages, devices, modules, etc., do not represent the order of precedence, and do not limit the "first" and "second" to be different types.
[0052] 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.
[0053] Figure 2 : is a flow chart of a method for predicting the prognosis risk of LUAD provided in the second aspect of an embodiment of the present invention. Specifically, the method comprises the following steps:
[0054] 101: Obtain gene expression data of the sample to be tested;
[0055] In some embodiments, the term "subject" or "test subject" or "test sample" used herein refers to any animal (e.g., mammal), including but not limited to humans, non-human primates, rodents, etc., which will be the recipient of a specific treatment. Generally, the terms "subject" and "patient" are used interchangeably herein when referring to human subjects. Preferably, the subject is a human.
[0056] In some embodiments, the sample to be tested is a patient undergoing clinical prognosis assessment.
[0057] 102: Inputting the gene expression data of the sample to be tested into the prognostic risk model disclosed in the first aspect of the present application to obtain any one of the classification results of the second IB-L, the second IB-M and the second IB-H;
[0058] In some embodiments, the method for constructing the prognostic risk model comprises:
[0059] S1, obtain the gene expression data of the IB stage training set samples and the corresponding survival status of the samples;
[0060] S2, based on the survival situation, cluster analysis is performed on the IB stage training set samples to obtain different subtypes including the first IB-L and the first IB-H and the gene expression situation characterizing each group of subtypes; optionally, the different subtypes and the gene expression situation characterizing each group of subtypes in S2 also include: the first IB-M subtype and the gene expression situation characterizing the subtype; optionally, the cluster analysis in S2 includes: clustering the IB stage training set samples according to the similarity of the gene expression data to obtain clustering results including any one or more of IB1, IB2, IB3 and IB4; subdividing the clustering results of any one or more of IB1, IB2, IB3 and IB4 into clustering results including any one or more of IB-L, IB-M and IB-H according to the survival situation of the IB stage training set samples; optionally, the direction of clustering the IB stage training set samples according to the similarity of the gene expression data includes: consistency clustering (CDF);
[0061] S3, screening the genes in the first IB-H subtype and the first IB-L subtype respectively by a machine learning algorithm to obtain the key genes of IB-H and IB-L respectively; optionally, the machine learning algorithm includes any one or more of the following: rf, gbm, XGBoost, LASSO;
[0062] S4, taking the intersection of the IB-H key gene and the IB-L key gene to obtain a prediction gene; constructing the prognostic risk model based on the prediction gene, including different subtypes of the second IB-L, the second IB-M and the second IB-H, and characterizing the gene expression of each subtype. Optionally, the prognostic risk model constructed based on the prediction gene is established based on logistic regression.
[0063] In some embodiments, the predicted genes include at least two of CENPE (centromere protein E), FSCN1 (fascin actin-bundling protein 1), and SLC2A1 (solute carrier family 2 member 1);
[0064] Optionally, the risk score of the predicted gene corresponding to the second IB-L is in the L interval, the risk score of the predicted gene corresponding to the second IB-M is in the M interval, and the risk score of the predicted gene corresponding to the second IB-H is in the H interval; that is, the risk scores of the predicted genes corresponding to the second IB-L, the second IB-M and the second IB-H are from low to high; in some embodiments, the ratio of the L interval, the M interval and the H interval for the three types of typing is the same as the ratio of the training set or the test set samples. If the ratio of the training set or the test set samples is d:e:f, the standard for defining the L interval, the M interval and the H interval corresponding to the three types of typing is also d:e:f.
[0065] Optionally, the L interval, M interval and H interval are obtained by training the prediction genes in the training set samples and the different subtypes of the samples;
[0066] Optionally, risk score = CENPE*a+FSCN1*b+SLC2A1*c, where a is the coefficient of CENPE, b is the coefficient of FSCN1, and c is the coefficient of SLC2A1. Optionally, a is 5.2249, b is 1.9510, and c is 2.3061, but the specific values of coefficients a / b / c vary with changes in training set samples.
[0067] In some embodiments, stage IB1 has the best prognosis and stage IB4 has the worst prognosis; the OS of patients in stage IB1 is similar to that in stage IA; the OS of patients in stage IB4 is much worse than that in stage IA, and there is no significant difference in the prognosis of stage IB2 compared with stage IA, but there is a trend towards worsening; the prognosis of stage IB3 is worse than that of stage IA, but the difference is not as significant as that of stage IB4, and the OS of stage IB4 is similar to that of stage II; the OS of patients with stage IB1, IB2, and IB3 diseases is longer than that of patients with stage II disease (especially patients with stage IB1 disease).
[0068] In some embodiments, IB-L (similar to stage IA, the clinical treatment strategy is surgery alone), IB-M (between stage IA and stage II, true clinical and biological stage IB, its treatment requires comprehensive consideration and active clinical trials), IB-H (worse than stage IA, similar to stage II, indicating the need for active adjuvant therapy after surgery);
[0069] Figure 3 is a schematic flow chart of a method provided by a third aspect of an embodiment of the present invention, wherein the method comprises:
[0070] 201, obtaining expression data of predicted genes in the sample to be tested; the predicted genes include any two or three of CENPE (centromere protein E), FSCN1 (fascin actin-bundling protein 1), and SLC2A1 (solute carrier family 2 member 1);
[0071] 202, calculating the expression levels of the at least two prediction genes, and classifying the sample to be tested according to the expression levels to obtain any one of the classification results of the second IB-L, the second IB-M, and the second IB-H;
[0072] Optionally, the predicted gene expressions are all negative and are defined as the second IB-L, the predicted gene expressions are all positive and are defined as the second IB-H, and non-second IB-L and non-IB-H are defined as the second IB-M.
[0073] In some embodiments, the survival period of the sample to be tested is predicted based on the classification result; based on the result of the second IB-L, a result indicating that the survival period of the sample to be tested is long is output; based on the result of the second IB-H, a result indicating that the survival period of the sample to be tested is short is output; based on the result of the second IB-M, a result indicating that the survival period of the sample to be tested is between the second IB-L and the second IB-H is output;
[0074] Optionally, the method also includes: based on the result of the second IB-L, outputting a result that the survival period of the sample to be tested is similar to stage IA; based on the result of the second IB-H, outputting a result that the survival period of the sample to be tested is similar to stage II; based on the result of the second IB-M, outputting a result that the survival period of the sample to be tested is between stage IA and stage II.
[0075] In some embodiments, the method further includes: based on the result of the second IB-L, outputting the result that the sample to be tested can refer to the result of the stage IA treatment strategy; based on the result of the second IB-H, outputting the result that the sample to be tested can refer to the result of the stage II treatment strategy; based on the result of the second IB-M, outputting the result that the sample to be tested needs to comprehensively consider the treatment mode.
[0076] In some embodiments, the second IB-L classification results show obvious metabolic activity characteristics; the second IB-H classification results show obvious active proliferation-related processes; and the second IB-M classification results show the aggregation of immune signaling pathways.
[0077] In some embodiments, the method further includes: based on the gene expression data, if RTK-RAS is shown to have mutated, obtaining a result that the sample to be tested is a second IB-L; if NOTCH is shown to have mutated, obtaining a result that the sample to be tested is a second IB-M; if TP53 and Wnt are shown to have mutated, obtaining a result that the sample to be tested is a second IB-H.
[0078] In the present invention, the prognostic indicators include objective response rate, overall survival rate (Overall survival rate, Overall survival, OS), progression-free survival (PFS), objective response rate (Objective Response Rate, ORR), time to progression (time to progress, TTP), disease-free survival (Disease-free survival, DFS), time to treatment failure (time to treatment failure, TTF), response rate (RR), complete response (CR), partial response (PR). The term "prognosis" is recognized in the art and includes predictions about possible disease processes or disease development, especially about the possibility of disease remission, disease recurrence, tumor recurrence, metastasis and death. "Good prognosis" means the possibility that a patient with cancer, especially pancreatic cancer, remains disease-free (i.e., cancer-free). "Poor prognosis" means the possibility of recurrence or recurrence, metastasis or death of a potential cancer or tumor.
[0079] Figure 4 is a schematic diagram of a computer device provided by an embodiment of the present invention, such as Figure 4 As shown, the device may include: one or more processors, and one or more memories; wherein the memories store computer-readable codes, and when the computer-readable codes are run by the one or more processors, the method described above may be executed.
[0080] The processor in this embodiment can be an integrated circuit chip with signal processing capabilities. The above processor can be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The disclosed methods, operations and logic block diagrams in the embodiments of the present disclosure can be implemented or executed. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc., which can be an X86 architecture or an ARM architecture.
[0081] In general, various example embodiments of the present disclosure may be implemented in hardware or dedicated circuits, software, firmware, logic, or any combination thereof. Certain aspects may be implemented in hardware, while other aspects may be implemented in firmware or software that may be executed by a controller, microprocessor, or other computing device. When various aspects of the disclosed embodiments are illustrated or described as block diagrams, flow charts, or using some other graphical representation, it will be understood that the blocks, devices, systems, techniques, or methods described herein may be implemented in hardware, software, firmware, dedicated circuits or logic, general purpose hardware or controllers or other computing devices, or some combination thereof as non-limiting examples.
[0082] For example, the method or device according to the embodiment of the present disclosure may also be implemented by Figure 5 The architecture of the computing device 3000 shown in FIG. Figure 5 As shown, the computing device 3000 may include a bus 3010, one or more CPUs 3020, a read-only memory (ROM) 3030, a random access memory (RAM) 3040, a communication port 3050 connected to a network, an input / output component 3060, a hard disk 3070, etc. The storage device in the computing device 3000, such as ROM 3030 or hard disk 3070, may store various data or files used for processing and / or communication of the method provided by the present disclosure and program instructions executed by the CPU. The computing device 3000 may also include a user interface 3080. Of course, Figure 5 The architecture shown is only exemplary and can be omitted according to actual needs when implementing different devices. Figure 5 One or more components of a computing device are shown.
[0083] The embodiment of the present invention also provides a computer-readable storage medium, such as Figure 6As shown, it is a schematic diagram of a storage medium provided by an embodiment of the present invention, and a computer readable instruction 4010 is stored on the computer storage medium 4020. When the computer readable instruction 4010 is executed by the processor, the method according to the embodiment of the present disclosure described with reference to the above figures can be executed. The computer readable storage medium in the embodiment of the present disclosure can be a volatile memory or a non-volatile memory, or can include both volatile and non-volatile memories. The non-volatile memory can be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM) or a flash memory. The volatile memory can be a random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of RAM are available, such as static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDRSDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM), and direct memory bus random access memory (DR RAM). It should be noted that the memory of the methods described herein is intended to include, but is not limited to, these and any other suitable types of memory. It should be noted that the memory of the methods described herein is intended to include, but is not limited to, these and any other suitable types of memory.
[0084] The embodiments of the present disclosure further provide a computer program product or system, including a computer program, which implements the steps of the above method when executed by a processor.
[0085] In some embodiments, this embodiment further discloses a system for constructing a prognostic risk model for stage IB LUAD, the system comprising:
[0086] The first acquisition module is used to obtain the gene expression data of the IB stage training set samples and the corresponding survival status of the samples;
[0087] An initial classification module, used for performing cluster analysis on the IB stage training set samples based on the survival status, to obtain different subtypes including the first IB-L and the first IB-H and the gene expression status characterizing each group of subtypes;
[0088] A key gene screening module, used to screen the genes in the first IB-H subtype and the first IB-L subtype respectively through a machine learning algorithm to obtain IB-H and IB-L key genes respectively;
[0089] The final classification module is used to take the intersection of the IB-H key gene and the IB-L key gene to obtain the prediction gene; based on the prediction gene, the prognostic risk model including different subtypes of the second IB-L, the second IB-M and the second IB-H and the gene expression of each subtype is constructed.
[0090] In some embodiments, this embodiment further discloses a system for predicting the prognosis risk of LUAD, the system comprising:
[0091] The second acquisition module is used to obtain gene expression data of the sample to be tested;
[0092] The first result output module is used to input the gene expression data of the sample to be tested into the prognostic risk model disclosed in the first aspect of the present application to obtain any one of the classification results of the second IB-L, the second IB-M and the second IB-H.
[0093] Optionally, the gene expression data includes any two or three of the following genes: CENPE, FSCN1, SLC2A1.
[0094] In some embodiments, another embodiment of the present application further discloses a system for predicting the prognosis risk of LUAD, the system comprising:
[0095] The third acquisition module is used to obtain the expression data of the predicted genes in the sample to be tested; the predicted genes include any two or three of CENPE (centromere protein E), FSCN1 (fascin actin-bundling protein 1), and SLC2A1 (solute carrier family 2 member 1);
[0096] The second result output module is used to calculate the expression levels of the at least two prediction genes, classify the test samples according to the expression levels, and obtain any one of the classification results of the second IB-L, the second IB-M and the second IB-H.
[0097] Optionally, the predicted gene expressions are all negative and are defined as the second IB-L, the predicted gene expressions are all positive and are defined as the second IB-H, and non-second IB-L and non-IB-H are defined as the second IB-M.
[0098] In some embodiments, the system further comprises:
[0099] A survival period output module is used to predict the survival period of the sample to be tested based on the classification result; based on the result of the second IB-L, output a result that the survival period of the sample to be tested is long; based on the result of the second IB-H, output a result that the survival period of the sample to be tested is short; based on the result of the second IB-M, output a result that the survival period of the sample to be tested is between the second IB-L and the second IB-H;
[0100] In some embodiments, the system further comprises:
[0101] The survival reference staging output module is used to output a result that the survival period of the sample to be tested is similar to stage IA based on the result of the second IB-L; to output a result that the survival period of the sample to be tested is similar to stage II based on the result of the second IB-H; and to output a result that the survival period of the sample to be tested is between stage IA and stage II based on the result of the second IB-M.
[0102] In some embodiments, the system further comprises:
[0103] The treatment mode output module is used to output the results of the treatment strategy for the stage IA for the samples to be tested based on the results of the second IB-L; to output the results of the treatment strategy for the stage II for the samples to be tested based on the results of the second IB-H; and to output the results of the treatment mode that needs to be comprehensively considered for the samples to be tested based on the results of the second IB-M.
[0104] Example:
[0105] 1. Methods:
[0106] 1.1 Dataset
[0107] Public bulk transcriptome datasets containing clinical information from the TCGA and GEO databases were downloaded (TCGA-LUAD, GSE11969, GSE13213, GSE26939, GSE30219, GSE31210, GSE41271, GSE42127, GSE50081, GSE63459, GSE68465, GSE68571, GSE72094, and GSE81089), and patients with stage IB, IA, or II LUAD with detailed clinical information were selected for analysis.
[0108] 1.2 Involved ideas
[0109] The design process of this study is as follows Figure 7As shown in the figure. First, we randomly divided 14 datasets (759IB, 909IA, and 482II) of 2150 LUAD patients into 2 datasets (dataset 1 for training; dataset 2 for testing) in a ratio of 6:4 according to stage stratification (stage IB, IA, and II). In dataset 1, 455 stage IB patients were clustered based on prognosis-related genes. The optimal number of clusters was selected according to the consistent cumulative distribution function (CDF), and the prognostic features of high-risk (IB-H), medium-risk (IB-M), and low-risk (IB-L) subtypes were further identified by prognostic features. Then, six machine learning algorithms were used to map the gene expression profiles and molecular types of the markers in dataset 1, thereby generating discriminant models to classify the molecular subtypes in dataset 2. Then, the clinical significance of the risk classification of stage IB was evaluated by comparing the prognosis of stage IB with that of stage IA and stage II; next, the biological phenotypes of IB-H, IB-M, and IB-L subtypes were evaluated by gene set enrichment analysis (GSEA), Reactome analysis, and gene mutation comparison. In addition, four machine learning methods were used to screen key genes and establish a risk score to stratify the risk of stage IB patients.
[0110] 1.3 Statistical methods
[0111] ConsensusClusterPlus software package was used for cluster analysis (maxK = 10, reps = 1000, pItem = 0.8, pFeature = 1, clusterAlg = "pam", distance = "Euclidean"). UMAP (UMAP package) and t-distributed stochastic neighbor embedding (t-SNE) (Rtsne package) were used to display the clustering results. Kaplan-Meier curves were generated to show the survival probability, and the log-rank test and weighted log-rank test (Peto-Peto test) were used for statistical analysis. The P values of multiple group comparisons were corrected by the Benjamini-Hochberg (BH) method to achieve an accurate balance between false positives and false negatives. The Cox proportional hazards regression model was used for risk factor assessment. The biological process enrichment was based on the Reactome database (R package: ReactomePA), and GSEA based on the Hallmark database was used to analyze different pathways (ClusterProfiler package); the TCGAmutations package and maftools package were used to analyze the mutation status. K nearest neighbor (knn), naive Bayes (nb), partial least squares (pls), stochastic gradient boosting (gbm), extreme gradient boosting (xgbTree), random forest (rf), and least absolute shrinkage and selection operator (LASSO) methods were used for machine learning on transcriptomic data (caret, randomForest, e1071, and glmnet packages). Logistic regression was used to build risk scores (glmnet package). Nomograms were used to visualize the probability predictions of risk scores.
[0112] 2. Results
[0113] 2.1 Molecular classification of stage IB LUAD
[0114] First, we integrated the gene expression data of 14 datasets by z-transformation. The unintegrated dataset showed obvious batch effects, but the integrated batch effects were not obvious. According to the stratification ratio of stage IA, IB, and II, we divided the patients into two datasets in a ratio of 6:4 (dataset 1 for training and dataset 2 for testing). Among the 455 stage IB patients in dataset 1, 1059 prognostic genes were identified by Cox univariate analysis (p < 0.1) ( Figure 8 A). We then performed consensus clustering of stage IB LUAD patients based on the similarity of gene expression profiles and divided the patients into four categories—stage IB1, stage IB2, stage IB3, and stage IB4—based on the approximate maximum and minimum increases in CDF ( Figure 8B). UMAP and t-SNE can display high-dimensional data of similar distances on a two-dimensional plane to visualize the distribution of different IB subtypes ( Figure 8 C). The prognostic differences among stages IB1, IB2, IB3, and IB4 were further compared, and significant differences in survival were found, with stage IB4 having the worst prognosis (median OS time of 44 months), stage IB1 having the best prognosis (median OS time was not observed at the end of follow-up), and stages IB3 (median OS time of 89 months) and IB2 (median OS time of 112 months) being in between ( Figure 8 D) These results suggest that stage IB LUAD is highly heterogeneous and further classification of stage IB-LUAD is needed, especially for precision treatment.
[0115] 2.2 Clinical significance of molecular classification of stage IB LUAD
[0116] We compared the molecular classification of stage IB patients with the prognostic differences of stage IA and stage II patients to evaluate the clinical characteristics of stage IB patients, thereby providing assistance for clinical precision treatment. We found that the OS of stage IB1 patients was similar to that of stage IA patients. This finding suggests that stage IA patients may have similar prognostic characteristics to stage IB1 patients and require similar clinical operations, that is, no need for postoperative adjuvant chemotherapy ( Fig. 9 A). However, the OS of stage IB4 patients was significantly worse than that of stage IA patients. This finding suggests that stage IB4 patients may require progressive surgery, such as postoperative adjuvant chemotherapy, rather than just surgical resection ( Fig. 9 A). In addition, although the prognosis of stage IB2 patients was not significantly different from that of stage IA patients, the prognosis of stage IB2 patients tended to be worse than that of stage IA patients. The prognosis of stage IB3 patients was worse than that of stage IA patients, but the difference was not as significant as that of stage IB4 patients ( Fig. 9 A). In addition, the OS of stage IB4 patients is similar to that of stage II patients, suggesting that the clinical treatment of stage IB4 patients can be similar to that of stage II patients, that is, adjuvant therapy should be performed promptly after surgery ( Fig. 9 B). OS was longer in patients with stage IB1, IB2, and IB3 disease than in patients with stage II disease (especially stage IB1 disease). Fig. 9 B). For the convenience of clinical use and combined with the above prognostic characteristics, IB1 is defined as a low-risk subtype (IB-L), IB2 and IB3 are defined as intermediate-risk subtypes (IB-M), and IB4 is defined as a high-risk subtype (IB-H). The prognosis of stage IB-L is similar to that of stage IA, but significantly better than that of stage II, which suggests that the clinical treatment strategy for stage IB-L patients should be similar to that of stage IA patients, that is, surgical treatment alone ( Fig. 9 C). The prognosis of stage IB-H is significantly worse than that of stage IA and similar to that of stage II, suggesting that active adjuvant therapy is required after surgery ( Fig. 9D). The prognosis of stage IB-M is between stage IA and stage II. Therefore, stage IB-M can be regarded as a true clinical and biological stage IB, and its treatment requires comprehensive consideration and active clinical trials ( Fig. 9 C, 9D).
[0117] 2.3 Establishment and testing of molecular classification model for stage IB LUAD
[0118] We used six machine learning algorithms (rf, gbm, xgbTree, knn, pls, and naive_bayes) to establish the above molecular typing discriminant model. Model construction and training were performed based on the classification in Dataset 1, and the model effect was tested in Dataset 2. In Dataset 1, the confusion matrix showed that all six algorithms obtained good classification ( Fig.10 A, 10D, 10G). In addition, in dataset 2, the above molecular classification has good value in guiding clinical decision-making, especially for the high-risk subtype (IB-H). All 6 algorithms showed that IB-H has a poor prognosis, similar to stage II. These findings suggest that an active clinical intervention strategy should be adopted for IB-H ( Fig.10 C, 10F, 10I). The low-risk (IB-L) and intermediate-risk (IB-H) subtypes showed excellent prognosis compared with stage II and similar prognosis compared with stage IA. However, the p-values of some models (compared with stage IA) were close to statistical significance (0.05), indicating that the classification effect of low-risk (IB-L) and intermediate-risk (IB-M) subtypes was not as good as that of high-risk subtype (IB-H) (there may be some confounding factors) ( Fig.10 B, 10E, 10H).
[0119] 2.4 Biological significance of molecular classification of stage IB LUAD
[0120] First, we evaluated the biological significance of the molecular subtypes in Dataset 1. Through GSEA based on gene expression features (from the hallmark database), we found that IB-L showed obvious metabolic activity characteristics, IB-M showed aggregation of immune signaling pathways, and IB-H showed obvious active proliferation-related processes. In Dataset 2, the molecular subtypes classified by the above 6 machine algorithms showed similar biological characteristics to Dataset 1. That is, IB-L showed metabolic activity characteristics, and IB-M involved the aggregation of immune signaling pathways. IB-H involved active proliferation-related processes. In summary, we identified IB-H as a proliferation subtype, IB-M as an immune subtype, and IB-L as a metabolic subtype.
[0121] 2.5 Establishing a risk score for LUAD IB molecular classification
[0122] Given the complexity of machine learning and the difficulty of distinguishing the above IB-L and IB-M subtypes, we attempted to establish a risk score for LUADIB molecular subtypes. In Dataset 1, four machine learning algorithms (rf, gbm, XGBoost, and LASSO) were used to screen the key genes identified by IB-H and IB-L ( Fig.11 AD). By cross-analysis of the most significant genes (top 30), we identified three signature genes (centromere protein E (CENPE), fascin actin bundling protein 1 (FSCN1), and solute carrier family 2 member 1 (SLC2A1)) ( Fig.11 E). The IB molecular classification risk score was established based on logistic regression (risk score = CENPE*(5.2249)+FSCN1*(1.9510)+SLC2A1*(2.3061)). Based on the proportion of IB molecular types in the above consensus clustering (dataset 1), patients with stage IB LUAD were further divided into IB-H group, IB-M group and IB-L group according to the risk score. The classification effect of the risk score was good. IB-L showed the best prognosis, which was similar to stage IA, but significantly better than patients with stage II disease, indicating that patients with stage IA disease and stage IB-L patients (surgical treatment only) can adopt similar clinical management strategies ( Fig.11 F). The prognosis of stage IB-H is the worst, similar to stage II but significantly higher than stage IA, suggesting that the treatment of stage IB-H disease can refer to the treatment of stage II disease (surgery and active adjuvant therapy). The treatment of stage IB-M disease is between stage IA and stage II disease patients, and the treatment model for this population needs to be comprehensively considered ( Fig.11 F). Moreover, we made a similar proportion of division in Dataset 2, and the molecular classification had the same clinical significance ( Fig.11 G). In summary, the risk score can classify stage IB LUAD well and can not only identify the high-risk subtype, but also effectively identify the low-risk subtype and the intermediate-risk subtype, which suggests that it can complement machine learning-based classification.
[0123] Clinical and biological significance of key genes in molecular classification of stage IB LUAD
[0124] We continued to explore the clinical and biological significance of the three key genes (CENPE, FSCN1, and SLC2A1) that constitute the risk score of stage IB LUAD. We found that CENPE and SLC2A1 were significantly overexpressed in tumor tissues in both datasets 1 and 2 ( Fig.12 A, 12B). However, there was no significant difference in FSCN1 between tumor tissues and normal tissues ( Fig.12A, 12B). Next, we analyzed the prognostic significance of the three genes, and CENPE showed a clear negative prognostic significance in dataset 1, but did not show prognostic significance in dataset 2 ( Fig.12 C). In dataset 1, FSCN1 also had a negative prognostic effect, but the p-value was close to 0.05. These results suggest that the prognostic prediction is unstable and is not significant in dataset 2 ( Fig.12 D) SLC2A1 had no prognostic significance in either dataset 1 or dataset 2 for stage IB LUAD ( Fig.12 E). So, what effect will the combination of these three genes have? We divided stage IB LUAD patients into high-risk group (CENPE+FSCN1+SLC2A1+), low-risk group (CENPE-FSCN1-SLC2A1-) and intermediate-risk group (other groups) according to the median expression levels of the three genes. In both datasets 1 and 2, the above classifications have obvious clinical significance. That is, the high-risk group (IB-H) has the worst prognosis, similar to stage II. The low-risk group (IB-L) has the best prognosis, similar to IA. The intermediate-risk group is between the two groups ( Fig.12 FI). GSEA showed that the above three genes in datasets 1 and 2 were closely associated with the malignant biological characteristics of tumors, including proliferation, epithelial-mesenchymal transition, and glycolytic pathway enrichment. In fact, the cancer-promoting effects of the above three genes have been reported in the literature, especially the proliferation function of CENPE, the metastasis function of FSCN1, and the regulatory function of SLC2A1 in tumor metabolism. In summary, we believe that CENPE, FSCN1, and SLC2A1 play a key role in the malignant progression of stage IB LUAD by regulating proliferation, metastasis, and metabolic activities, and their combination can be used as an indicator for prognosis and risk classification of stage IB LUAD.
[0125] It should be noted that the flowcharts and block diagrams in the accompanying drawings illustrate the possible architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each box in the flowchart or block diagram can represent a module, a program segment, or a part of a code, and the module, program segment, or a part of the code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of boxes in the block diagram and / or flowchart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.
[0126] In general, various example embodiments of the present disclosure may be implemented in hardware or dedicated circuits, software, firmware, logic, or any combination thereof. Certain aspects may be implemented in hardware, while other aspects may be implemented in firmware or software that may be executed by a controller, microprocessor, or other computing device. When various aspects of the disclosed embodiments are illustrated or described as block diagrams, flow charts, or using some other graphical representation, it will be understood that the blocks, devices, systems, techniques, or methods described herein may be implemented in hardware, software, firmware, dedicated circuits or logic, general purpose hardware or controllers or other computing devices, or some combination thereof as non-limiting examples.
[0127] 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.
[0128] 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.
[0129] 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.
[0130] 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.
[0131] The exemplary embodiments of the present disclosure described in detail above are merely illustrative and not restrictive. It should be understood by those skilled in the art that various modifications and combinations may be made to these embodiments or their features without departing from the principles and spirit of the present disclosure, and such modifications should fall within the scope of the present disclosure.
Claims
1. A method for constructing a prognostic risk model for stage IB LUAD, characterized in that: The method comprises: S1, obtain the gene expression data of the IB stage training set samples and the corresponding survival status of the samples; S2, performing cluster analysis on the IB stage training set samples based on the survival status to obtain different subtypes including the first IB-L and the first IB-H and the gene expression status characterizing each subtype; S3, screening the genes in the first IB-H subtype and the first IB-L subtype by a machine learning algorithm to obtain IB-H and IB-L key genes, respectively; S4, taking the intersection of the IB-H key gene and the IB-L key gene to obtain the predicted gene; constructing the different subtypes including the second IB-L, the second IB-M and the second IB-H and the prognostic risk model characterizing the gene expression of each group of subtypes based on the predicted gene.
2. The method for constructing a prognostic risk model for stage IB LUAD according to claim 1, characterized in that: The different subtypes in S2 and the gene expression conditions representing each group of subtypes also include: the first IB-M subtype and the gene expression conditions representing the subtype; Optionally, the cluster analysis in S2 includes: clustering the IB stage training set samples according to the similarity of the gene expression data to obtain clustering results including any one or more of IB1, IB2, IB3 and IB4; subdividing any one or more of the clustering results of IB1, IB2, IB3 and IB4 into any one or more of the clustering results of IB-L, IB-M and IB-H according to the survival status of the IB stage training set samples; Optionally, the machine learning algorithm includes any one or more of the following: rf, gbm, XGBoost, LASSO; Optionally, the method of clustering the samples of the stage IB training set according to the similarity of the gene expression data includes: consistency clustering; Optionally, the prognostic risk model constructed based on the predicted genes is established based on logistic regression.
3. The method for constructing a prognostic risk model for stage IB LUAD according to claim 1, characterized in that: The predicted genes include at least two of CENPE, FSCN1, and SLC2A1; Optionally, the risk score of the predicted gene corresponding to the second IB-L is in the L interval, the risk score of the predicted gene corresponding to the second IB-M is in the M interval, and the risk score of the predicted gene corresponding to the second IB-H is in the H interval; that is, the risk scores of the predicted genes corresponding to the second IB-L, the second IB-M and the second IB-H are from low to high; Optionally, the L interval, M interval and H interval are obtained by training the prediction genes in the training set samples and the different subtypes of the samples; Optionally, risk score = CENPE*a+FSCN1*b+SLC2A1*c, where a is the coefficient of CENPE, b is the coefficient of FSCN1, and c is the coefficient of SLC2A1.
4. A method for predicting the prognosis risk of LUAD, characterized in that: The method comprises: Obtaining gene expression data of the sample to be tested; Inputting the gene expression data of the sample to be tested into the prognostic risk model of any one of claims 1 to 3 to obtain any one of the classification results of the second IB-L, the second IB-M and the second IB-H; Optionally, the gene expression data includes any two or three of the following genes: CENPE, FSCN1, SLC2A1.
5. A method for predicting the prognosis risk of LUAD, characterized in that: The method comprises: Acquiring expression data of predicted genes in the sample to be tested; the predicted genes include any two or three of CENPE, FSCN1, and SLC2A1; Calculating the expression levels of the at least two prediction genes, and classifying the samples to be tested according to the expression levels to obtain any one of the classification results of the second IB-L, the second IB-M, and the second IB-H; Optionally, the predicted gene expressions are all negative and are defined as the second IB-L, the predicted gene expressions are all positive and are defined as the second IB-H, and non-second IB-L and non-IB-H are defined as the second IB-M.
6. The method for predicting the prognosis risk of LUAD according to claim 4 or 5, characterized in that: Predicting the survival period of the sample to be tested based on the classification result; outputting a result indicating that the survival period of the sample to be tested is long based on the result of the second IB-L; and outputting a result indicating that the survival period of the sample to be tested is short based on the result of the second IB-H; Based on the result of the second IB-M, outputting the result that the survival period of the sample to be tested is between the second IB-L and the second IB-H; Optionally, the method further comprises: based on the result of the second IB-L, outputting a result that the survival period of the sample to be tested is similar to stage IA; Based on the result of the second IB-H, the output is a result that the survival period of the sample to be tested is similar to that of stage II; based on the result of the second IB-M, the output is a result that the survival period of the sample to be tested is between stage IA and stage II.
7. The method for predicting the prognosis risk of LUAD according to claim 4 or 5, characterized in that: The method further includes: based on the result of the second IB-L, outputting the result of the treatment strategy of stage IA for the sample to be tested; based on the result of the second IB-H, outputting the result of the treatment strategy of stage II for the sample to be tested; Based on the results of the second IB-M, the output of the samples to be tested needs to comprehensively consider the results of the treatment model.
8. A computer device, characterized in that: The device comprises: a memory and a processor; the memory is used to store a computer program; the processor executes the computer program to implement the steps of the method according to any one of claims 1 to 7.
9. A computer-readable storage medium, characterized in that: A computer program is stored thereon, and when the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.
10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method described in any one of claims 1 to 7 are implemented.
Citation Information
Patent Citations
Prognosis evaluation system for patients with early-stage lung adenocarcinoma (LUAD) and application thereof
CN111394456A
Colon cancer prognosis marker gene and screening, prognosis prediction and model construction method thereof
CN115394445A
Method for constructing lung adenocarcinoma immune prognosis model based on gender
CN116189775A
Lung adenocarcinoma prognosis evaluation system and device
CN117954097A
