Method and system for constructing autophagy-based acute myeloid leukemia prognosis model

Abstract

The present application relates to the technical field of medical model construction, and in particular to a method and system for constructing an acute myeloid leukemia prognosis model based on autophagy. The life state, KPS score and reads quantity of each gene of historical AML patients are obtained; the KPS score of each AML patient at different follow-up times, the reads quantity of each gene and the life state of different AML patients are combined to determine the gene expression coefficient of different genes, and the genes are divided into autophagy-related genes and non-autophagy-related genes; the probability of the non-autophagy-related genes being FAML-related genes is determined; the probability of the non-autophagy-related genes being FAML-related genes is corrected, and FAML-related genes are screened out; the proportional hazards regression model is corrected in combination with the influence of the FAML-related genes on the autophagy-related genes, and a prognosis risk model of the patient is constructed. The present application improves the prediction accuracy of the model by correcting the proportional hazards regression model.

Description

Technical Field

[0001] This invention relates to the field of medical model construction technology, specifically to a method and system for constructing a prognostic model for acute myeloid leukemia based on autophagy. Background Technology

[0002] Acute myeloid leukemia (AML) is a highly heterogeneous hematopoietic malignancy with significant differences in clinical prognosis, influenced by multiple factors including gene mutations, epigenetic regulation, and the immune microenvironment. In recent years, the role of autophagy in the occurrence, development, and treatment response of AML has received increasing attention. Autophagy not only affects the survival and drug resistance of AML cells but may also regulate the cell death mechanism induced by chemotherapy drugs. Therefore, the construction of AML prognostic models based on autophagy-related genes (ARGs) has significant clinical value, providing precise guidance for personalized treatment and survival prediction. Currently, although AML prognostic models based on gene mutations and cell signaling pathways exist, precise survival prediction methods incorporating autophagy regulatory mechanisms are still lacking, necessitating further improvements in the predictive accuracy and clinical applicability of AML prognosis.

[0003] Current prognostic models for acute myeloid leukemia (AML) are mostly based on single somatic mutations, analyzing the impact of different gene mutations on survival, but they fail to adequately consider the interactions of familial AML (FAML)-related genes under different risk gene backgrounds. FAML-related genes may affect the oncogenicity of certain high-risk genes, such as FLT3-ITD and TP53, or enhance the treatment response of certain low-risk genes (such as NPM1 and CEBPA) by regulating key biological processes such as autophagy, gene repair, and cell differentiation. Therefore, traditional proportional hazards regression models are unable to quantify the corrective effect of FAML-related genes on survival prediction, resulting in low model accuracy and difficulty in providing accurate personalized prognostic assessments. Summary of the Invention

[0004] To address the technical problem of low model accuracy caused by the inability of traditional proportional hazards regression models to quantify the corrective effects of FAML-related genes on survival prediction, this invention aims to provide a method and system for constructing an autophagy-based prognostic model for acute myeloid leukemia. The specific technical solution adopted is as follows: In a first aspect, embodiments of the present invention provide a method for constructing a prognostic model for acute myeloid leukemia based on autophagy, the method comprising: Obtain the vital signs, KPS scores, and number of reads for each gene of historical AML patients; By combining the correlation between the KPS score and the number of reads of each gene at different follow-ups for each AML patient, the probability that each gene in each AML patient is a protective gene is determined; based on the probability that different genes in different AML patients are protective genes and the life status of different AML patients, the gene expression coefficients of different genes are determined; based on the gene expression coefficients, genes are divided into autophagy-related genes and non-autophagy-related genes. By combining the correlation between the number of reads of autophagy-related genes and the correlation between gene expression coefficients, the probability that a non-autophagy-related gene is a FAML-related gene is determined; based on the number of gene reads, the probability that a non-autophagy-related gene is a FAML-related gene is adjusted, and FAML-related genes are screened out. By combining the effects of FAML-related genes on autophagy-related genes, the proportional hazards regression model was modified to construct a prognostic risk model for patients.

[0005] Furthermore, the determination of the probability that each gene in each AML patient is a protective gene by combining the correlation between the KPS score and the number of reads for each gene at different follow-ups includes: Any AML patient was used as the target AML patient; for the target AML patient, the difference in KPS score between adjacent follow-ups was calculated as the degree of change in KPS score of the target AML patient; Using any gene as the target gene, the difference in the number of reads of the target gene in adjacent follow-up visits of the target AML patient is calculated as the degree of change in the number of target genes in the target AML patient. Calculate the correlation coefficients of score change sequences and number change sequences across all follow-ups, and use this as the probability that the target gene for the target AML patient is a protective gene.

[0006] Furthermore, determining the gene expression coefficients of different genes based on the probability that different genes are protective genes in different AML patients and the life status of different AML patients includes: Compare the vital signs of the currently analyzed AML patients with all AML patients to determine the gene expression intensity of the currently analyzed AML patients; Based on the probability that different genes are protective genes in different AML patients and the gene expression intensity in different AML patients, the gene expression coefficients of different genes are determined.

[0007] Furthermore, the comparison of the vital status of the currently analyzed AML patient with that of all AML patients, and the determination of the gene expression intensity of the currently analyzed AML patient, includes: Obtain the mean prognostic survival time for the acute myeloid leukemia type in the currently analyzed AML patients; The difference between the prognostic survival time and the prognostic mean survival time of the AML patients at the last follow-up in the current analysis was used as the numerator, and the standard deviation of the prognostic survival time of all AML patients under the same acute myeloid leukemia type in the current analysis was used as the denominator. The ratio of the numerator and denominator was used as the gene expression intensity of the AML patients in the current analysis.

[0008] Furthermore, determining the gene expression coefficients of different genes based on the probability that different genes in different AML patients are protective genes and the gene expression intensity in different AML patients includes: Target AML patients with any AML and target genes with any gene; The product of the gene expression intensity of the target AML patient and the probability that the target gene of the target AML patient is a protective gene is used as the patient expression coefficient of the target gene of the target AML patient. The sum of the patient expression coefficients of the target gene in all AML patients is taken as the gene expression coefficient of the target gene.

[0009] Furthermore, the determination of the probability that a non-autophagy-related gene is a FAML-related gene by combining the correlation between the number of reads of autophagy-related genes and the correlation between gene expression coefficients includes: For each gene, the gene expression level vector is composed of the number of gene reads from different patients at the same follow-up. Using any non-autophagy-related gene as the non-autophagy-related gene to be analyzed, the absolute value of the correlation coefficient between the gene expression level vector of the non-autophagy-related gene to be analyzed and each autophagy-related gene is taken as the first correlation probability; the ratio of the gene expression coefficient of each autophagy-related gene to the gene to be analyzed is taken as the second correlation probability; the product of the first correlation probability and the second correlation probability is taken as the comprehensive correlation probability between the non-autophagy-related gene to be analyzed and each autophagy-related gene. The sum of the overall correlation probabilities between the non-autophagy-related gene to be analyzed and all autophagy-related genes is taken as the probability that the non-autophagy-related gene to be analyzed is a FAML-related gene.

[0010] Furthermore, the step of adjusting the probability of non-autophagy-related genes becoming FAML-related genes based on the number of gene reads, and screening for FAML-related genes, includes: For each gene, the gene fluctuation value is determined based on the fluctuation of the number of gene reads at different follow-up visits; By combining the probability that non-autophagy-related genes are FAML-related genes with gene fluctuation values, the corrected probability that non-autophagy-related genes are FAML-related genes is obtained. The non-autophagy-related genes with a pre-set number of corrected probabilities are considered as FAML-related genes.

[0011] Furthermore, the proportional hazards regression model is modified by incorporating the effects of FAML-related genes on autophagy-related genes to construct a prognostic risk model for patients, including: The autophagy-related gene corresponding to the maximum combined correlation probability between FAML-related genes and each autophagy-related gene is taken as the FAML-related influencing gene; The proportional hazards regression model was modified by combining the number of reads of autophagy-related genes with the number of reads of FAML-related genes.

[0012] Furthermore, the proportional hazards regression model is modified by combining the number of reads of autophagy-related genes with the number of reads of FAML-related influencing genes, including: ; in, For AML patients at time t, let X be the risk function with covariate X; exp is the exponential function with the natural constant as the base; m is the number of autophagy-related genes. denoted as the number of reads of the p-th autophagy-related gene at time t; is the regression coefficient of the interaction term corresponding to the p-th autophagy-related gene; X is the covariate; For a linear combination of covariates X; Baseline risk function; Let q be the number of reads for the q-th FAML-related gene at time t; The number of reads of the FAML-related influencing genes corresponding to the q-th FAML-related gene at time t; is the regression coefficient of the cross term corresponding to the q-th FAML-related gene.

[0013] Secondly, an autophagy-based prognostic model construction system for acute myeloid leukemia is provided, the system comprising the following modules: The data acquisition module is used to acquire the vital signs, KPS scores, and number of reads for each gene of historical AML patients. The judgment module is used to determine the probability that each gene in each AML patient is a protective gene by combining the correlation between the KPS score and the number of reads of each gene at different follow-ups; based on the probability that different genes in different AML patients are protective genes and the life status of different AML patients, the gene expression coefficient of different genes is determined; based on the gene expression coefficient, the genes are divided into autophagy-related genes and non-autophagy-related genes. The determination module is used to combine the correlation between the number of reads of autophagy-related genes and the correlation between gene expression coefficients to determine the probability that a non-autophagy-related gene is a FAML-related gene; based on the number of gene reads, the probability that a non-autophagy-related gene is a FAML-related gene is adjusted, and FAML-related genes are screened out. The model correction module is used to combine the influence of FAML-related genes on autophagy-related genes to correct the proportional hazards regression model and construct a prognostic risk model for patients.

[0014] Thirdly, embodiments of the present invention provide an electronic device, including a memory and a processor, wherein the memory stores executable code, and when the processor executes the executable code, it implements the various possible implementations of the first aspect.

[0015] Fourthly, embodiments of the present invention provide a computer program product comprising: computer program code, which, when executed on a computer, causes the computer to perform the method described in the first aspect or any possible implementation thereof.

[0016] Fifthly, embodiments of the present invention provide a computer-readable storage medium having a computer program stored thereon, which, when executed in a computer, causes the computer to perform the various possible implementations of the first aspect.

[0017] The embodiments of the present invention have at least the following beneficial effects: This invention analyzes different gene data of AML patients and their KPS scores at different follow-up visits to determine the probability of a gene being a protective gene, and then analyzes the patient's survival time. When the prognostic survival time of AML patients differs significantly from the average survival time, it may be due to strong expression of protective genes, enhancing anti-cancer ability and resulting in longer survival, or it may be due to strong expression of risk genes, promoting cancer progression and resulting in shorter survival. Therefore, this invention analyzes the patient's life status, determines the expression coefficients of different genes, and combines the probability of a gene being a protective gene with the gene expression coefficients to ultimately determine the gene expression intensity. This allows for the classification of autophagy-related genes and non-autophagy-related genes, analyzes the impact of familial acute myeloid leukemia (FAML)-related genes on autophagy gene expression, and considers the interaction between FAML-related genes and AML gene mutations. This modifies the traditional proportional hazards regression model, effectively improving the accuracy and reliability of prognostic prediction and enhancing the predictive precision of the risk model. Attached Figure Description

[0018] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art 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 effort.

[0019] Figure 1 A flowchart illustrating a method for constructing a prognostic model for acute myeloid leukemia based on autophagy, provided in one embodiment of the present invention. Figure 2 A system block diagram of an autophagy-based prognostic model construction system for acute myeloid leukemia provided in one embodiment of the present invention; Figure 3 This is a schematic diagram of the structure of a computer device provided in one embodiment of the present invention. Detailed Implementation

[0020] To further illustrate the technical means and effects of the present invention in achieving its intended purpose, the following detailed description, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of the method and system for constructing an autophagy-based prognostic model for acute myeloid leukemia proposed in this invention.

[0021] In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments may be combined in any suitable form.

[0022] In the description of the embodiments of the present invention, unless otherwise stated, " / " means "or". For example, A / B can mean A or B. The "and / or" in the text is merely a description of the relationship between related objects, indicating that there can be three relationships. For example, A and / or B can mean: A exists alone, A and B exist simultaneously, and B exists alone. In addition, in the description of the embodiments of the present invention, "multiple" means two or more.

[0023] Hereinafter, the terms "first" and "second" are used for descriptive purposes only and should not be construed as implying or suggesting relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature.

[0024] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0025] The embodiments of the present invention will now be described with reference to the accompanying drawings. Those skilled in the art will recognize that, with technological advancements and the emergence of new scenarios, the technical solutions provided by the embodiments of the present invention are also applicable to similar technical problems.

[0026] The following description, in conjunction with the accompanying drawings, details the specific scheme of the method and system for constructing an autophagy-based prognostic model for acute myeloid leukemia provided by this invention.

[0027] Please see Figure 1 The diagram illustrates a flowchart of a method for constructing a prognostic model for acute myeloid leukemia based on autophagy, according to an embodiment of the present invention. The method includes the following steps: Step S100: Obtain the vital signs, KPS scores, and number of reads for each gene of historical AML patients.

[0028] To analyze the potential role of autophagy in the occurrence, progression, and treatment response of AML, gene expression profiles of historical AML patients were extracted from public databases and clinical samples, and genes closely related to the autophagy process and autophagy-related gene families were screened. Data collection covered patient populations with different molecular subtypes, treatment statuses, and survival outcomes to ensure the comprehensiveness and representativeness of the study.

[0029] The follow-up data for individual historical AML patients includes follow-up record number, follow-up time and number of follow-ups, patient survival status, patient vital signs, number of days from diagnosis to death if the patient has died, whether there has been a relapse, gene expression level, Karnofsky score (KPS), number of gene reads, etc. The number of reads of a gene obtained by sequencing is a well-known technique in the art and will not be described in detail here.

[0030] Then, based on the gene expression data of historical AML patients during multiple historical follow-up periods, gene screening was performed to predict the prognostic risk of AML.

[0031] Step S200: Combine the correlation between the KPS score and the number of reads of each gene in each AML patient at different follow-up visits to determine the probability that each gene in each AML patient is a protective gene; determine the gene expression coefficient of different genes based on the probability that different genes in different AML patients are protective genes and the life status of different AML patients; and classify the genes into autophagy-related genes and non-autophagy-related genes based on the gene expression coefficient.

[0032] First, we distinguish between autophagy-related genes and non-autophagy-related genes.

[0033] Considering that the physical condition of patients with a history of AML may improve or worsen between two consecutive follow-ups, the difference in KPS scores and gene expression between two consecutive follow-ups can be used to better determine the impact of differences in gene expression on the patient's condition.

[0034] Any AML patient was used as the target AML patient; for the target AML patient, the difference in KPS score between adjacent follow-ups was calculated as the degree of change in KPS score of the target AML patient.

[0035] For example, taking the i-th historical AML patient as the target AML patient, let the i-th be the target AML patient. The number of historical AML patients in the first The KPS score at the next follow-up was , No. The number of historical AML patients in the first The KPS score at the next follow-up was Then, the change in KPS score between two consecutive follow-up visits is .

[0036] Using any gene as the target gene, the difference in the number of reads of the target gene between adjacent follow-up visits in the target AML patient is calculated as the degree of change in the number of target genes in the target AML patient.

[0037] For example, taking the i-th historical AML patient as the target AML patient and the p-th gene as the target gene, let the i-th gene be the target gene. The number of historical AML patients in the first At the next follow-up, the number of reads corresponding to the p-th gene was: , No. The number of reads corresponding to the p-th gene in a patient with a history of AML at the u-1 follow-up is: Then, in two consecutive follow-up visits, the change in the number of gene reads is: It should be noted that in subsequent steps, patients with a history of AML will be referred to as AML patients.

[0038] If the p-th gene has a high correlation with the patient's condition, then the corresponding gene is expressed more strongly. The higher the patient's condition score, the higher the likelihood that the gene is a protective gene for the patient.

[0039] Calculate the correlation coefficients of score change sequences and number change sequences across all follow-ups, and use this as the probability that the target gene for the target AML patient is a protective gene.

[0040] Among them, the rating change sequence is constructed by the rating change in chronological order, and the quantity change sequence is constructed by the quantity change in chronological order.

[0041] In this embodiment of the invention, the correlation coefficient between the score change series and the quantity change series under all follow-up periods is calculated using the Pearson correlation coefficient.

[0042] In this embodiment of the invention, the expression of a gene is characterized by the number of gene reads. When the expression of a corresponding gene increases in a patient, the KPS also increases, indicating that the gene has a certain protective effect on the patient. Conversely, the probability of the corresponding risk gene is higher.

[0043] After determining the probability that each gene in each AML patient is a protective gene, the gene expression coefficients of different genes are further determined based on the probability that different genes in different AML patients are protective genes and the life status of different AML patients.

[0044] Because acute myeloid leukemia (AML) has many subtypes and precursor tumors, distinguished by morphological, immunophenotypic, cytochemical, and genetic abnormalities, different types of the disease significantly impact prognosis and treatment. The WHO (World Health Organization) classification system divides it into seven categories. Within different categories, the average lifespan of patients also varies. This analysis aims to obtain the average prognostic survival time for the specific AML type in the current analysis of AML patients. For example, let the average prognostic survival time for the i-th AML patient be denoted as... .

[0045] For a single patient, obtaining the first The prognostic survival time of a patient up to the last follow-up, if the first... If a patient has died at the last follow-up, the survival time at the last follow-up is taken as the prognostic survival time and recorded as follows: Otherwise, the prognostic survival time at the last follow-up visit, as updated in real time, is recorded as... Based on prognostic survival time and prognostic mean survival time The differences between patients were analyzed to determine the strength of expression of the corresponding genes.

[0046] When the Prognostic survival time and average prognostic survival time for each patient When there is a significant difference between the two, it may be due to the stronger expression of protective genes, which enhances the anti-cancer ability and makes the patient's survival time longer, but it may also be due to the stronger expression of risk genes, which promotes the progression of cancer and makes the patient's survival time shorter.

[0047] Therefore, by comparing the current AML patient's life status with that of all AML patients, the gene expression intensity of the current AML patient was determined.

[0048] The difference between the prognostic survival time and the prognostic mean survival time of the AML patients at the last follow-up in the current analysis was used as the numerator, and the standard deviation of the prognostic survival time of all AML patients under the same acute myeloid leukemia type in the current analysis was used as the denominator. The ratio of the numerator and denominator was used as the gene expression intensity of the AML patients in the current analysis.

[0049] In some embodiments, taking the i-th AML patient as an example of the currently analyzed AML patient, the gene expression intensity corresponding to the i-th AML patient is... The calculation formula is: ;in, denoted as the standard deviation of prognostic survival time for all AML patients of the same acute myeloid leukemia type as the i-th AML patient.

[0050] For genes with high gene expression intensity, the reliability of their corresponding protective or risk genes is relatively high. Therefore, further, based on the probability that different genes in different AML patients are protective genes and the gene expression intensity in different AML patients, the gene expression coefficients of different genes are determined.

[0051] The product of the gene expression intensity of the target AML patient and the probability that the target gene of the target AML patient is a protective gene is used as the patient expression coefficient of the target gene in the target AML patient; the sum of the patient expression coefficients of the target genes of all AML patients is used as the gene expression coefficient of the target gene.

[0052] In some embodiments, the p-th gene is used as the target gene, and the gene expression coefficient of the p-th gene is... The calculation formula is: Where N is the number of AML patients; Let p be the probability that the p-th gene in the i-th AML patient is a protective gene.

[0053] Finally, based on the gene expression coefficients, genes are divided into autophagy-related genes and non-autophagy-related genes: genes with expression coefficients greater than a preset threshold are classified as autophagy-related genes, and a set of autophagy-related genes is formed from these genes. Genes with expression coefficients less than or equal to a preset threshold are considered non-autophagy-related genes, and a set of non-autophagy-related genes is formed from these genes. In this embodiment of the invention, the preset proportion threshold is the minimum value among the top 20% of gene expression coefficients from largest to smallest. This can also be understood as sorting the gene expression coefficients from largest to smallest and selecting... For the front Genes that exhibit certain characteristics are considered autophagy-related genes, while those that do not are considered non-autophagy-related genes.

[0054] Step S300: Combine the correlation between the number of reads of autophagy-related genes and non-autophagy-related genes and the correlation between gene expression coefficients to determine the probability that a non-autophagy-related gene is a FAML-related gene; adjust the probability that a non-autophagy-related gene is a FAML-related gene based on the number of gene reads, and screen out FAML-related genes.

[0055] Because familial AML may have a recessive genetic susceptibility, even patients with low known risk gene mutation burdens may still have short survival. For example, certain FAML-related genes, such as RUNX1 or GATA2, are highly enriched in familial AML, and even if patients do not exhibit traditionally high-risk mutations, their AML progression may still be rapid. Therefore, we consider introducing FAML-related genes as correction factors to enhance the accuracy of AML prognosis prediction. In this embodiment of the invention, we first analyze the follow-up data and gene expression coefficients of autophagy-related and non-autophagy-related genes to screen for FAML-related genes.

[0056] First, the probability of FAML-related genes was calculated based on cross-sectional comparisons among AML patients.

[0057] When predicting patient survival, some FAML-related genes may inhibit or amplify autophagy-related genes, leading to lower accuracy in predicting prognostic risk. Therefore, we consider combining the correlation of gene expression coefficients between different AML patients to identify some possible FAML-related genes.

[0058] Considering the magnitude of the correlation between gene expression in two different parts, when the set of non-autophagy-related genes... Collection of genes related to autophagy When the absolute value of the correlation between some genes is strong, there is a strong mutual influence between them, that is... Some genes in the middle will affect The genes in the middle can be enhanced or suppressed.

[0059] For each gene, the gene expression level vector is composed of the number of gene reads from different patients at the same follow-up.

[0060] To obtain the gene expression level vector for each gene, it's important to note that the elements in the gene expression level vector correspond to the same patient order; that is, the elements are sorted according to the patient's sequence number. Let the expression level vector of the autophagy-related gene p in multiple patients be denoted as... The correlation coefficient between the expression level vectors of autophagy-related gene p1 and non-autophagy-related gene q2 is... ,in , .

[0061] Using any non-autophagy-related gene as the non-autophagy-related gene to be analyzed, the absolute value of the correlation coefficient between the gene expression level vector of the non-autophagy-related gene to be analyzed and each autophagy-related gene is taken as the first correlation probability; the ratio of the gene expression coefficient of each autophagy-related gene to the gene to be analyzed is taken as the second correlation probability; and the product of the first correlation probability and the second correlation probability is taken as the comprehensive correlation probability between the non-autophagy-related gene to be analyzed and each autophagy-related gene.

[0062] In some embodiments, non-autophagy-related gene q2 is used as the non-autophagy-related gene to be analyzed, and the combined correlation probability between non-autophagy-related gene q2 and autophagy-related gene p1 is... The calculation formula is: ;in The gene expression coefficient of p1, an autophagy-related gene; The gene expression coefficient of q2, a non-autophagy-related gene; This represents the second relevant probability; This represents the first relevant probability.

[0063] The sum of the overall correlation probabilities between the non-autophagy-related gene to be analyzed and all autophagy-related genes is taken as the probability that the non-autophagy-related gene to be analyzed is a FAML-related gene.

[0064] Then, based on longitudinal comparisons of multiple historical tests in a single AML patient, the probability of non-autophagy-related genes being FAML-related genes was corrected to obtain FAML-related genes.

[0065] Between germline mutations and somatic mutations, germline mutations exist in all cells, are inherited from parents or occur at the fertilized egg stage, and usually do not disappear throughout an individual's life, while somatic mutations exist only in specific cells, such as leukemia cells, and do not appear in all normal cells. They may disappear or be added as the disease progresses or is treated, such as when leukemia is in remission.

[0066] Considering the stable occurrence of FAML-related genes in individual patients, the probability of non-autophagy-related genes being FAML-related genes was adjusted.

[0067] Based on the number of gene reads, the probability of non-autophagy-related genes becoming FAML-related genes is adjusted, and FAML-related genes are screened out.

[0068] For each gene, the gene fluctuation value is determined based on the fluctuation in the number of gene reads at different follow-up visits. In this embodiment of the invention, for each patient, the standard deviation of the number of gene reads at different follow-up visits is used as the gene fluctuation value.

[0069] By combining the probability that a non-autophagy-related gene is a FAML-related gene with the gene fluctuation value, the corrected probability that a non-autophagy-related gene is a FAML-related gene is obtained; among them, the probability that a non-autophagy-related gene is a FAML-related gene is directly proportional to the corrected probability, and the gene fluctuation value is directly proportional to the corrected probability.

[0070] In some embodiments, each gene can obtain the probability and gene fluctuation value of a non-autophagy-related gene being a FAML-related gene at each follow-up, calculate the ratio of the probability of a non-autophagy-related gene being a FAML-related gene to the gene fluctuation value as the follow-up probability, and take the maximum follow-up probability of each gene in multiple follow-ups as the corrected probability of the gene.

[0071] In descending order of numerical value, the first preset number of non-autophagy-related genes with corrected probabilities are identified as FAML-related genes. In this embodiment, the preset number is 5; in other embodiments, the implementer may adjust this value according to actual circumstances.

[0072] In step S400, the proportional hazards regression model is modified by combining the influence of FAML-related genes on autophagy-related genes, and a prognostic risk model for patients is constructed.

[0073] The genes that primarily influence patient prognosis and survival time are the identified FAML-related genes. The traditional Cox survival prediction model, also known as a proportional hazards regression model, is used as the base model for model refinement. First, the patient's time at time... The corresponding proportional hazards regression model is: ; in, For the proportional hazards regression model, the risk function for AML patients at time t, with covariate X; This represents the baseline risk function, indicating the individual's risk when all covariates X are zero. The regression coefficients of the variables; The number of reads for a gene; denoted as the number of genes; exp is an exponential function with the natural constant as the base. It should be noted that the basic risk function for an individual is obtained using the Breslow estimation method, which is existing technology for those skilled in the art and will not be elaborated upon here.

[0074] Considering the selected FAML-related genes, and given the interaction between these genes and their corresponding FAML-related influencing genes, the interaction term should be taken into account during Cox regression analysis. Therefore, the corrected prognostic risk model for the patient at time t is denoted as follows. for: ; in, For AML patients at time t, let X be the risk function with covariate X; exp is the exponential function with the natural constant as the base; m is the number of autophagy-related genes. denoted as the number of reads of the p-th autophagy-related gene at time t; is the regression coefficient of the variable corresponding to the p-th autophagy-related gene; X is the covariate; For a linear combination of covariates X; Baseline risk function; Let q be the number of reads for the q-th FAML-related gene at time t; The number of reads of the FAML-related influencing genes corresponding to the q-th FAML-related gene at time t; is the regression coefficient of the cross term corresponding to the q-th FAML-related gene.

[0075] Based on the improved prognostic risk model, the model parameters are determined. The number of gene reads and the survival status of AML patients are input into the prognostic risk model. The survival status of AML patients corresponds to a binary variable. When the survival status is 1, it indicates that a death event has occurred. When the survival status is 0, it indicates that the patient has not died. The prognostic risk model outputs regression coefficients, including regression coefficients of gene variables and regression coefficients of interaction terms.

[0076] In a preferred embodiment of the present invention, for AML patients, relevant data of the patients to be predicted are collected, including the number of reads of AML-related genes and protective or risk genes. Missing values ​​are processed, such as by mean imputation or multiple imputation, to ensure that the variable form is consistent with the trained Cox regression model. The Cox regression model is used to calculate the patient's risk score, predict the patient's survival probability at different time points (e.g., 1 year, 3 years, 5 years), and Kaplan-Meier survival curves are plotted. The following are schematic diagrams of survival curves for some patients.

[0077] Please see Figure 2 The diagram illustrates a system block diagram of an autophagy-based prognostic model construction system for acute myeloid leukemia according to an embodiment of the present invention. The system includes the following modules: The data acquisition module is used to acquire the vital signs, KPS scores, and number of reads for each gene of historical AML patients. The judgment module is used to determine the probability that each gene in each AML patient is a protective gene by combining the correlation between the KPS score and the number of reads of each gene at different follow-ups; based on the probability that different genes in different AML patients are protective genes and the life status of different AML patients, the gene expression coefficient of different genes is determined; based on the gene expression coefficient, the genes are divided into autophagy-related genes and non-autophagy-related genes. The determination module is used to combine the correlation between the number of reads of autophagy-related genes and the correlation between gene expression coefficients to determine the probability that a non-autophagy-related gene is a FAML-related gene; based on the number of gene reads, the probability that a non-autophagy-related gene is a FAML-related gene is adjusted, and FAML-related genes are screened out. The model correction module is used to combine the influence of FAML-related genes on autophagy-related genes to correct the proportional hazards regression model and construct a prognostic risk model for patients.

[0078] Alternatively, the transmission medium may be a wired link, such as, but not limited to, coaxial cable, fiber optic cable and digital subscriber line, or a wireless link, such as, but not limited to, wireless Fidelity (WIFI), Bluetooth and mobile device networks.

[0079] It should be noted that the device provided in the above embodiments is only an example of the division of the above functional modules. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the computer device can be divided into different functional modules to complete all or part of the functions described above.

[0080] Figure 3 This is a schematic diagram of the structure of a computer device provided in an embodiment of the present invention. For example, as shown... Figure 3 As shown, the computer device 500 includes: a memory 510, a processor 520, and a computer program 530 stored in the memory 510 and running on the processor 520, wherein when the processor 520 executes the computer program 530, the computer device can execute any of the autophagy-based prognostic model construction methods for acute myeloid leukemia described above.

[0081] Furthermore, embodiments of the present invention also protect an apparatus that may include a memory and a processor, wherein the memory stores executable program code, and the processor is used to call and execute the executable program code to execute the method for constructing an autophagy-based prognostic model for acute myeloid leukemia provided in embodiments of the present invention.

[0082] In this embodiment of the invention, the device can be divided into functional modules according to the above method example. For example, each module can correspond to a separate function, or two or more functions can be integrated into one processing module. The integrated module can be implemented in hardware. It should be noted that the module division in this embodiment is illustrative and is only a logical functional division. In actual implementation, there may be other division methods.

[0083] When each module is divided according to its function, the device may also include a signal uploading module, a determination module, and an adjustment module. It should be noted that all relevant content of each step involved in the above method embodiments can be referenced from the functional descriptions of the corresponding functional modules, and will not be repeated here.

[0084] It should be understood that the apparatus provided in this embodiment of the invention is used to execute the above-described method for constructing an autophagy-based prognostic model for acute myeloid leukemia, and thus can achieve the same effect as the above-described implementation method.

[0085] When using integrated units, the device may include a processing module and a storage module. When applied to a device, the processing module can be used to control and manage the device's operations. The storage module can be used to support the device in executing program code, etc. The processing module may be a processor or a controller, which can implement or execute various exemplary logic blocks, modules, and circuits as described in this disclosure. The processor may also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of Digital Signal Processing (DSP) and a microprocessor, etc., and the storage module may be a memory.

[0086] In addition, the device provided in the embodiments of the present invention may specifically be a chip, component or module. The chip may include a connected processor and a memory. The memory is used to store instructions. When the processor calls and executes the instructions, the chip can execute the method for constructing an autophagy-based prognostic model for acute myeloid leukemia provided in the above embodiments.

[0087] This invention also provides a computer-readable storage medium storing computer program code. When the computer program code is run on a computer, the computer executes the aforementioned method steps to implement the autophagy-based prognostic model construction method for acute myeloid leukemia provided in the above embodiments.

[0088] This invention also provides a computer program product that, when run on a computer, causes the computer to perform the aforementioned steps to implement the method for constructing an autophagy-based prognostic model for acute myeloid leukemia provided in the above embodiments.

[0089] In this invention, the apparatus, computer-readable storage medium, computer program product, or chip provided in the embodiments are all used to execute the corresponding methods described above. Therefore, the beneficial effects they achieve can be referred to the beneficial effects in the corresponding methods described above, and will not be repeated here. Through the above description of the embodiments, those skilled in the art will understand that, for the sake of convenience and brevity, only the division of the above functional modules is used as an example. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. In the embodiments provided by this invention, it should be understood that the disclosed apparatus and method can be implemented in other ways.

[0090] The device embodiments described above are merely illustrative. For example, the division of modules or units is only a logical functional division. In actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another device, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0091] It should also be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or terminal device that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or terminal device. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or terminal device that includes said element.

[0092] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0093] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

[0094] The above content is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the protection scope of the present invention.

Claims

1. A method for constructing a prognostic model for acute myeloid leukemia based on autophagy, characterized in that, The method includes the following steps: Obtain the vital signs, KPS scores, and number of reads for each gene of historical AML patients; By combining the correlation between the KPS score and the number of reads of each gene at different follow-ups for each AML patient, the probability that each gene in each AML patient is a protective gene is determined; based on the probability that different genes in different AML patients are protective genes and the life status of different AML patients, the gene expression coefficients of different genes are determined; based on the gene expression coefficients, genes are divided into autophagy-related genes and non-autophagy-related genes. By combining the correlation between the number of reads of autophagy-related genes and the correlation between gene expression coefficients, the probability that a non-autophagy-related gene is a FAML-related gene is determined; based on the number of gene reads, the probability that a non-autophagy-related gene is a FAML-related gene is adjusted, and FAML-related genes are screened out. By combining the effects of FAML-related genes on autophagy-related genes, the proportional hazards regression model was modified to construct a prognostic risk model for patients.

2. The method for constructing a prognostic model for acute myeloid leukemia based on autophagy according to claim 1, characterized in that, The method of determining the probability that each gene in each AML patient is a protective gene by combining the correlation between the KPS score and the number of reads for each gene at different follow-ups includes: Any AML patient was used as the target AML patient; for the target AML patient, the difference in KPS score between adjacent follow-ups was calculated as the degree of change in KPS score of the target AML patient; Using any gene as the target gene, the difference in the number of reads of the target gene in adjacent follow-up visits of the target AML patient is calculated as the degree of change in the number of target genes in the target AML patient. Calculate the correlation coefficients of score change sequences and number change sequences across all follow-ups, and use this as the probability that the target gene for the target AML patient is a protective gene.

3. The method for constructing a prognostic model for acute myeloid leukemia based on autophagy according to claim 1, characterized in that, The determination of gene expression coefficients for different genes based on the probability that different genes are protective genes in different AML patients and the life status of different AML patients includes: Compare the vital signs of the currently analyzed AML patients with all AML patients to determine the gene expression intensity of the currently analyzed AML patients; Based on the probability that different genes are protective genes in different AML patients and the gene expression intensity in different AML patients, the gene expression coefficients of different genes are determined.

4. The method for constructing a prognostic model for acute myeloid leukemia based on autophagy according to claim 3, characterized in that, The comparison of the vital signs of the currently analyzed AML patients with all AML patients, and the determination of the gene expression intensity of the currently analyzed AML patients, includes: Obtain the mean prognostic survival time for the acute myeloid leukemia type in the currently analyzed AML patients; The difference between the prognostic survival time and the prognostic mean survival time of the AML patients at the last follow-up in the current analysis was used as the numerator, and the standard deviation of the prognostic survival time of all AML patients under the same acute myeloid leukemia type in the current analysis was used as the denominator. The ratio of the numerator and denominator was used as the gene expression intensity of the AML patients in the current analysis.

5. The method for constructing a prognostic model for acute myeloid leukemia based on autophagy according to claim 3, characterized in that, The determination of gene expression coefficients for different genes based on the probability that different genes are protective genes in different AML patients and the gene expression intensity in different AML patients includes: Target AML patients with any AML and target genes with any gene; The product of the gene expression intensity of the target AML patient and the probability that the target gene of the target AML patient is a protective gene is used as the patient expression coefficient of the target gene of the target AML patient. The sum of the patient expression coefficients of the target gene in all AML patients is taken as the gene expression coefficient of the target gene.

6. The method for constructing a prognostic model for acute myeloid leukemia based on autophagy according to claim 1, characterized in that, The method of determining the probability that a non-autophagy-related gene is a FAML-related gene by combining the correlation between the number of reads of autophagy-related genes and the correlation between gene expression coefficients includes: For each gene, the gene expression level vector is composed of the number of gene reads from different patients at the same follow-up. Using any non-autophagy-related gene as the non-autophagy-related gene to be analyzed, the absolute value of the correlation coefficient between the gene expression level vector of the non-autophagy-related gene to be analyzed and each autophagy-related gene is taken as the first correlation probability; the ratio of the gene expression coefficient of each autophagy-related gene to the gene to be analyzed is taken as the second correlation probability; the product of the first correlation probability and the second correlation probability is taken as the comprehensive correlation probability between the non-autophagy-related gene to be analyzed and each autophagy-related gene. The sum of the overall correlation probabilities between the non-autophagy-related gene to be analyzed and all autophagy-related genes is taken as the probability that the non-autophagy-related gene to be analyzed is a FAML-related gene.

7. The method for constructing a prognostic model for acute myeloid leukemia based on autophagy according to claim 1, characterized in that, The process involves adjusting the probability of non-autophagy-related genes becoming FAML-related genes based on the number of gene reads, and then screening for FAML-related genes, including: For each gene, the gene fluctuation value is determined based on the fluctuation of the number of gene reads at different follow-up visits; By combining the probability that non-autophagy-related genes are FAML-related genes with gene fluctuation values, the corrected probability that non-autophagy-related genes are FAML-related genes is obtained. The non-autophagy-related genes with a pre-set number of corrected probabilities are considered as FAML-related genes.

8. The method for constructing a prognostic model for acute myeloid leukemia based on autophagy according to claim 6, characterized in that, The influence of FAML-related genes on autophagy-related genes was incorporated to modify the proportional hazards regression model, and a prognostic risk model for patients was constructed, including: The autophagy-related gene corresponding to the maximum combined correlation probability between FAML-related genes and each autophagy-related gene is taken as the FAML-related influencing gene; The proportional hazards regression model was modified by combining the number of reads of autophagy-related genes with the number of reads of FAML-related genes.

9. The method for constructing a prognostic model for acute myeloid leukemia based on autophagy according to claim 8, characterized in that, The proportional hazards regression model is modified by combining the number of reads of autophagy-related genes with the number of reads of FAML-related influencing genes, including: ； in, For AML patients at time t, let X be the risk function with covariate X; exp is the exponential function with the natural constant as the base; m is the number of autophagy-related genes. denoted as the number of reads of the p-th autophagy-related gene at time t; is the regression coefficient of the interaction term corresponding to the p-th autophagy-related gene; X is the covariate; For a linear combination of covariates X; Baseline risk function; Let q be the number of reads for the q-th FAML-related gene at time t; The number of reads of the FAML-related influencing genes corresponding to the q-th FAML-related gene at time t; is the regression coefficient of the cross term corresponding to the q-th FAML-related gene.

10. A system for constructing a prognostic model for acute myeloid leukemia based on autophagy, characterized in that, The system includes the following modules: The data acquisition module is used to acquire the vital signs, KPS scores, and number of reads for each gene of historical AML patients. The judgment module is used to determine the probability that each gene in each AML patient is a protective gene by combining the correlation between the KPS score and the number of reads of each gene at different follow-ups; based on the probability that different genes in different AML patients are protective genes and the life status of different AML patients, the gene expression coefficient of different genes is determined; based on the gene expression coefficient, the genes are divided into autophagy-related genes and non-autophagy-related genes. The determination module is used to combine the correlation between the number of reads of autophagy-related genes and the correlation between gene expression coefficients to determine the probability that a non-autophagy-related gene is a FAML-related gene; based on the number of gene reads, the probability that a non-autophagy-related gene is a FAML-related gene is adjusted, and FAML-related genes are screened out. The model correction module is used to combine the influence of FAML-related genes on autophagy-related genes to correct the proportional hazards regression model and construct a prognostic risk model for patients.

Images