A prognostic model of lysosome-related genes in hepatocellular carcinoma and its construction method and application

Abstract

The present invention discloses a prognostic model of lysosome-related genes for hepatocellular carcinoma, a construction method and an application thereof, wherein the lysosome-related genes include: ATG10, ATP6V0B, HMOX1, RAB7A, SQSTM1, TMEM41B, TP53, FTH1, RNF19B, RAMP3, TPP1, AP1M2, CLTA and DNAJC6. The present invention uses 14 lysosome-related genes to establish a prognostic model and related application products that can predict the overall survival rate of hepatocellular carcinoma patients, and verifies that the model has good predictive performance and has potential clinical application value in individualized hepatocellular carcinoma prognosis evaluation.

Description

Technical Field

[0001] The present invention belongs to the field of biomedicine, and in particular relates to a hepatocellular carcinoma lysosome-related gene prognosis model and a construction method and application thereof. Background Art

[0002] Hepatocellular carcinoma (HCC) accounts for about 75% to 80% of all primary liver cancers and is one of the most common primary liver cancers. The incidence and mortality of HCC are increasing year by year. At present, surgery is the main treatment method in clinical practice, combined with multidisciplinary comprehensive treatment methods such as interventional therapy, radiotherapy and chemotherapy, targeted drugs and immunotherapy. Although treatment technology has made great progress, the prognosis of patients is relatively poor due to factors such as recurrence and metastasis, and the 5-year survival rate is less than 18%. Since TNM staging can only evaluate the tumor status from a macroscopic level, its application in prognosis has certain limitations. However, with the improvement of medical level and progress in fields such as genetic testing and molecular targeted therapy, molecular-level prognostic evaluation provides new opportunities for customized individualized liver cancer treatment strategies and prognosis prediction.

[0003] Lysosomes are dynamic organelles involved in the metabolic decomposition of eukaryotic cells, and they mainly play a role in degrading macromolecules in cells. Lysosomes can not only degrade exogenous substances engulfed by cells, but also decompose cytoplasmic components produced by autophagosomes, such as damaged organelles and misfolded macromolecules. Under certain conditions, lysosomes can mediate programmed cell death, cell membrane repair, and tissue remodeling by releasing acid hydrolases into the cytoplasm or outside the cell. Studies have found that the expression level and activity of lysosomal acid hydrolases change to varying degrees in many tumors, and these changes are closely related to the clinical pathological process of tumors. In addition, increasing evidence shows that lysosomes regulate tumor invasion and metastasis, angiogenesis, and apoptosis or non-apoptotic cell death by releasing various enzymes. However, the relationship between lysosome-related genes and HCC prognosis has not been fully studied. Therefore, it is necessary to analyze the expression and prognosis of lysosome-related genes in HCC, and the development of a prognostic model based on lysosome-related genes may help predict the prognosis of HCC patients. Summary of the invention

[0004] In view of the deficiencies of the prior art, the present invention provides a hepatocellular carcinoma lysosome-related gene prognostic model and a construction method and application thereof, which are helpful for predicting the prognosis of HCC patients.

[0005] The present invention is achieved through the following technical solutions:

[0006] A method for constructing a lysosome-related gene prognostic model for hepatocellular carcinoma comprises the following steps:

[0007] Step 1) Download a training set related to hepatocellular carcinoma from the Cancer Genome Atlas database, which contains standardized RNA-seq data sets and clinical data, and select hepatocellular carcinoma patients with a survival time of more than 90 days and survival status and clinical information as the training set; download a standardized RNA-seq data set and clinical data of hepatocellular carcinoma samples from the International Cancer Genome Consortium database as a validation set;

[0008] Step 2) Obtain lysosome-related genes from the molecular signature database and use Cox regression analysis to perform prognostic analysis on lysosome-related genes, with P < 0.05 as the cutoff value for screening prognostic-related genes;

[0009] Step 3) using the initial candidate prognosis-related lysosomal genes, LASSO regression analysis was performed in the training set to screen out genes for constructing a risk score model;

[0010] The risk score is expressed as:

[0011]

[0012] Where n is the number of prognostic genes, exp i is the expression value of gene i, β i is the regression coefficient of gene i;

[0013] Step 4) Use the coefficient of each gene to calculate an individualized risk score, and divide HCC patients into high-risk and low-risk groups using the median risk score; use ROC to calculate the area under the curve (AUC) at multiple time points to evaluate the discriminatory ability of the prognostic model; finally, use the same risk score formula and critical value in the validation set to verify the accuracy of the model.

[0014] Preferably, the lysosome-related genes include: ATG10, ATP6V0B, HMOX1, RAB7A, SQSTM1, TMEM41B, TP53, FTH1, RNF19B, RAMP3, TPP1, AP1M2, CLTA, DNAJC6;

[0015] The risk scoring model is as follows:

[0016] Risk score = (0.1771 × ATG10 gene expression level) + (0.2816 × ATP6V0B gene expression level) + (0.0735 × HMOX1 gene expression level) + (0.198 × RAB7A gene expression level) + (0.0305 × SQSTM1 gene expression level) + (0.1693 × TMEM41B gene expression level) + (-0.0066 × TP53 gene expression level) + (-0.0746 × FTH1 gene expression level) + (0.037 × RNF19B gene expression level) + (0.1362 × RAMP3 gene expression level) + (0.0504 × TPP1 gene expression level) + (0.0122 × AP1M2 gene expression level) + (0.1078 × CLTA gene expression level) + (0.0062 × DNAJC62 gene expression level).

[0017] Preferably, the "ATG10" refers to a nucleic acid encoding all or part of the ATG10 protein or substantially identical to all or part of the nucleic acid sequence or its analog, and its Gene ID is 83734;

[0018] The "ATP6V0B" refers to a nucleic acid encoding all or part of the ATP6V0B protein or substantially identical to all or part of the nucleic acid sequence or its analog, and its Gene ID is 533;

[0019] The "HMOX1" refers to a nucleic acid encoding all or part of the HMOX1 protein or substantially identical to all or part of the nucleic acid sequence or its analog, and its Gene ID is 3162;

[0020] The "RAB7A" refers to a nucleic acid encoding all or part of the RAB7A protein or substantially identical to all or part of the nucleic acid sequence or an analog thereof, and its Gene ID is 7879;

[0021] The "SQSTM1" refers to a nucleic acid encoding all or part of the SQSTM1 protein or substantially identical to all or part of the nucleic acid sequence or its analog, and its Gene ID is 8878;

[0022] The "TMEM41B" refers to a nucleic acid encoding all or part of the TMEM41B protein or substantially identical to all or part of the nucleic acid sequence or an analog thereof, and its Gene ID is 440026;

[0023] The "TP53" refers to a nucleic acid encoding all or part of the TP53 protein or substantially identical to all or part of the nucleic acid sequence or its analog, and its Gene ID is 7157;

[0024] The "FTH1" refers to a nucleic acid encoding all or part of the FTH1 protein or substantially identical to all or part of the nucleic acid sequence or its analog, and its Gene ID is 2495;

[0025] The "RNF19B" refers to a nucleic acid encoding all or part of the RNF19B protein or substantially identical to all or part of the nucleic acid sequence or an analog thereof, and its Gene ID is 127544;

[0026] The "RAMP3" refers to a nucleic acid encoding all or part of the RAMP3 protein or substantially identical to all or part of the nucleic acid sequence or its analog, and its Gene ID is 10268;

[0027] The "TPP1" refers to a nucleic acid encoding all or part of the TPP1 protein or substantially identical to all or part of the nucleic acid sequence or its analog, and its Gene ID is 1200;

[0028] The "AP1M2" refers to a nucleic acid encoding all or part of the AP1M2 protein or substantially identical to all or part of the nucleic acid sequence or its analog, and its Gene ID is 10053;

[0029] The "CLTA" refers to a nucleic acid encoding all or part of the CLTA protein or substantially identical to all or part of the nucleic acid sequence or its analog, and its Gene ID is 1211;

[0030] The “DNAJC6” refers to a nucleic acid encoding all or part of the DNAJC6 protein or being approximately identical to all or part of the nucleic acid sequence or its analogs, and its Gene ID is 9829.

[0031] The prognostic model constructed by the above construction method.

[0032] The above-mentioned prognostic model is used in the preparation of a product for predicting the prognosis of hepatocellular carcinoma, and the detection sample for the expression amount of the characteristic gene is tumor tissue.

[0033] Preferably, the product is a kit or a chip.

[0034] A computer device comprises a memory and a processor, wherein the memory stores a computer program and the processor executes the following steps:

[0035] Step A) obtaining expression data of lysosome-related genes of hepatocellular carcinoma patients to be tested;

[0036] Step B) inputting the expression data of lysosome-related genes of the hepatocellular carcinoma patient to be tested into the above-mentioned prognostic model to calculate the risk score of the hepatocellular carcinoma patient to be tested;

[0037] Step C) predicting the overall survival rate of patients with hepatocellular carcinoma based on the risk score.

[0038] A computer readable storage medium having a computer program stored thereon, wherein the computer program is executed by a processor to:

[0039] Step A) obtaining expression data of lysosome-related genes of hepatocellular carcinoma patients to be tested;

[0040] Step B) inputting the expression data of lysosome-related genes of the hepatocellular carcinoma patient to be tested into the above-mentioned prognostic model to calculate the risk score of the hepatocellular carcinoma patient to be tested;

[0041] Step C) predicting the overall survival rate of patients with hepatocellular carcinoma based on the risk score.

[0042] A computer program product, comprising a computer program, wherein the computer program is executed by a processor to:

[0043] Step A) obtaining expression data of lysosome-related genes of hepatocellular carcinoma patients to be tested;

[0044] Step B) inputting the expression data of lysosome-related genes of the hepatocellular carcinoma patient to be tested into the above-mentioned prognostic model to calculate the risk score of the hepatocellular carcinoma patient to be tested;

[0045] Step C) predicting the overall survival rate of patients with hepatocellular carcinoma based on the risk score.

[0046] Compared with the prior art, the present invention has the following advantages:

[0047] (1) Based on the study of the relationship between lysosome-related genes and OS of patients with hepatocellular carcinoma, the present invention constructed a prognostic model containing 14 genes. Through ROC curve analysis, the AUC of the model was 0.794, 0.736 and 0.735 in the 1-year, 3-year and 5-year follow-up, respectively. Compared with traditional clinical prognostic indicators such as T, N, and M Stage, the model showed obvious advantages in predicting patient prognosis.

[0048] (2) Compared with the traditional TNM staging prognostic system, the present invention can more accurately predict the prognosis of patients with hepatocellular carcinoma. In addition, the prognostic prediction performance of the model has been verified.

[0049] (3) By detecting the expression levels of characteristic genes of patients with hepatocellular carcinoma, the model constructed by the present invention can calculate the patient's risk score and combine it with the nomogram to more intuitively evaluate the patient's 3-year and 5-year survival probability, which is impossible to achieve with the TNM staging prognostic system. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1The forest plot is for the univariate regression analysis of 54 genes;

[0051] Figure 2 To develop a prognostic model based on lysosome-related genes in the training set: A is the distribution diagram of candidate gene coefficients; B is the optimal lambda value parameter diagram; C is the distribution diagram of hepatocellular carcinoma risk score and survival status based on the prognostic model; D is the survival curve diagram of high- and low-risk groups; E is the time-dependent ROC curve of the prognostic model constructed with 14 lysosome-related genes;

[0052] Figure 3 To verify the prognostic model developed based on lysosome-related genes in the validation set: A is the distribution of risk scores and survival status of hepatocellular carcinoma based on the prognostic model; B is the survival curve of high-risk and low-risk groups; C is the time-dependent ROC curve of the prognostic model constructed by 14 lysosome-related genes;

[0053] Figure 4 Univariate (A) and multivariate (B) analyses of prognostic factors;

[0054] Figure 5 To construct a nomogram for survival prediction;

[0055] Figure 6 Calibration curve showing the accuracy of the nomogram. DETAILED DESCRIPTION

[0056] The present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0057] Example 1

[0058] A method for constructing a lysosome-related gene prognostic model for hepatocellular carcinoma, the specific steps are as follows:

[0059] 1. Data download and preprocessing

[0060] The training set related to hepatocellular carcinoma was downloaded from The Cancer Genome Atlas (TCGA), which contains standardized RNA-seq datasets (mRNA in FPKM format) and clinical data. Hepatocellular carcinoma patients with survival time > 90 days and survival status and clinical information were screened, with a total of 371 samples. The validation set was downloaded from the International Cancer Genome Consortium (IGCG) database, which contains standardized RNA-seq datasets (mRNA expression information) and clinical data of 240 hepatocellular carcinoma samples. Lysosome-related genes (LRGs) were downloaded from the Molecular Feature Database (MsigDB v7.4), and a total of 454 LRGs were downloaded.

[0061] 2. Screening of prognosis-related lysosome-related genes

[0062] Based on the TCGA dataset, LRGs obtained from the molecular signature database were identified. Cox regression analysis was used to perform prognostic analysis on LRGs. Genes with a P value of < 0.05 were considered to have an impact on the survival of patients with hepatocellular carcinoma and could be used as a cutoff value for screening prognosis-related genes. A total of 54 genes were identified ( Figure 1 ).

[0063] 3. LASSO Cox analysis

[0064] For the 54 genes identified, LASSO regression analysis was performed to screen variables. First, the coefficient of each candidate gene was calculated ( Figure 2 A), and then determine the optimal lambda value parameter ( Figure 2 In B), such as Figure 2 As shown in B, 14 genes were finally determined to constitute the prognostic gene signature for constructing a risk scoring model.

[0065] The risk score of each sample was calculated according to the formula. The risk score distribution and survival status of hepatocellular carcinoma based on the prognostic model were as follows Figure 2 As shown in C, the genes finally screened and determined for constructing the risk scoring model include the following 14 genes: ATG10, ATP6V0B, HMOX1, RAB7A, SQSTM1, TMEM41B, TP53, FTH1, RNF19B, RAMP3, TPP1, AP1M2, CLTA, and DNAJC6.

[0066] The "ATG10" refers to a nucleic acid encoding all or part of the ATG10 protein or substantially identical to all or part of the nucleic acid sequence or its analog, and its Gene ID is 83734;

[0067] The "ATP6V0B" refers to a nucleic acid encoding all or part of the ATP6V0B protein or substantially identical to all or part of the nucleic acid sequence or its analog, and its Gene ID is 533;

[0068] The "HMOX1" refers to a nucleic acid encoding all or part of the HMOX1 protein or substantially identical to all or part of the nucleic acid sequence or its analog, and its Gene ID is 3162;

[0069] The "RAB7A" refers to a nucleic acid encoding all or part of the RAB7A protein or substantially identical to all or part of the nucleic acid sequence or an analog thereof, and its Gene ID is 7879;

[0070] The "SQSTM1" refers to a nucleic acid encoding all or part of the SQSTM1 protein or substantially identical to all or part of the nucleic acid sequence or its analog, and its Gene ID is 8878;

[0071] The "TMEM41B" refers to a nucleic acid encoding all or part of the TMEM41B protein or substantially identical to all or part of the nucleic acid sequence or an analog thereof, and its Gene ID is 440026;

[0072] The "TP53" refers to a nucleic acid encoding all or part of the TP53 protein or substantially identical to all or part of the nucleic acid sequence or its analog, and its Gene ID is 7157;

[0073] The "FTH1" refers to a nucleic acid encoding all or part of the FTH1 protein or substantially identical to all or part of the nucleic acid sequence or its analog, and its Gene ID is 2495;

[0074] The "RNF19B" refers to a nucleic acid encoding all or part of the RNF19B protein or substantially identical to all or part of the nucleic acid sequence or an analog thereof, and its Gene ID is 127544;

[0075] The "RAMP3" refers to a nucleic acid encoding all or part of the RAMP3 protein or substantially identical to all or part of the nucleic acid sequence or its analog, and its Gene ID is 10268;

[0076] The "TPP1" refers to a nucleic acid encoding all or part of the TPP1 protein or substantially identical to all or part of the nucleic acid sequence or its analog, and its Gene ID is 1200;

[0077] The "AP1M2" refers to a nucleic acid encoding all or part of the AP1M2 protein or substantially identical to all or part of the nucleic acid sequence or its analog, and its Gene ID is 10053;

[0078] The "CLTA" refers to a nucleic acid encoding all or part of the CLTA protein or substantially identical to all or part of the nucleic acid sequence or its analog, and its Gene ID is 1211;

[0079] The “DNAJC6” refers to a nucleic acid encoding all or part of the DNAJC6 protein or being approximately identical to all or part of the nucleic acid sequence or its analogs, and its Gene ID is 9829.

[0080] A risk scoring model is constructed, that is, the calculation formula of the risk score is as follows:

[0081]

[0082] Where: n is the number of prognostic genes, exp i is the expression value of gene i, β i is the regression coefficient of gene i.

[0083] Risk score = (0.1771 × ATG10 gene expression level) + (0.2816 × ATP6V0B gene expression level) + (0.0735 × HMOX1 gene expression level) + (0.198 × RAB7A gene expression level) + (0.0305 × SQSTM1 gene expression level) + (0.1693 × TMEM41B gene expression level) + (-0.0066 × TP53 gene expression level) + (-0.0746 × FTH1 gene expression level) + (0.037 × RNF19B gene expression level) + (0.1362 × RAMP3 gene expression level) + (0.0504 × TPP1 gene expression level) + (0.0122 × AP1M2 gene expression level) + (0.1078 × CLTA gene expression level) + (0.0062 × DNAJC62 gene expression level).

[0084] The results of survival analysis showed that the survival time of patients in the high-risk group was significantly shorter than that in the low-risk group ( Figure 2 In order to evaluate the accuracy of the prognostic model composed of 14 genes in predicting the prognosis of hepatocellular carcinoma, 1-year, 3-year, and 5-year receiver operating characteristic (ROC) curve analysis was performed to compare the respective AUC values. The results showed that the risk score of hepatocellular carcinoma patients in the training cohort was significantly correlated with overall survival (OS) (P < 0.001). The ROC curve analysis showed that the AUCs for 1, 3, and 5 years were 0.794, 0.736, and 0.735, respectively ( Figure 2 Middle E). The AUC value showed that the prognostic model composed of 14 genes had good discriminatory performance for the prognosis of patients with hepatocellular carcinoma.

[0085] 4. Test set validation of the prognostic model’s performance in distinguishing the prognosis of patients with hepatocellular carcinoma

[0086] 240 HCC samples downloaded from the ICGC database were used as the test set. The risk score of each sample was calculated according to the formula obtained in the training set. The distribution of HCC risk scores and survival status based on the prognostic model are shown in Figure 2. Figure 3 As shown in A; Figure 3 As shown in Figure B, the risk score of patients with hepatocellular carcinoma was significantly correlated with OS (P < 0.05). Figure 3 As shown in Figure C, the AUC values ​​for 1 year, 3 years, and 5 years are 0.711, 0.720, and 0.540, respectively, indicating that the prediction performance has been verified in the test set.

[0087] 5. Risk score as an independent prognostic factor in patients with hepatocellular carcinoma

[0088] To further explore whether the risk score can serve as an independent clinical prognostic factor, univariate and multivariate Cox proportional hazards regression analyses were applied to the TCGA cohort.

[0089] like Figure 4 As shown in the figure, in the TCGA cohort, univariate analysis ( Figure 4 A) found that risk score and T stage were risk factors for overall survival in patients with hepatocellular carcinoma. Multivariate analysis ( Figure 4 Middle B) found that the risk score was still an independent prognostic factor for overall survival in HCC (HR = 2.492 [1.713-3.625], P < 0.001). Therefore, the risk score can be used as an independent prognostic factor for HCC patients.

[0090] 6. Nomogram predicts prognosis of hepatocellular carcinoma samples

[0091] Based on the independent prognostic factors obtained above, a nomogram was constructed ( Figure 5 ) and the calibration curve of the nomogram was drawn ( Figure 6 ) to predict the survival probability of patients with hepatocellular carcinoma.

[0092] like Figure 6 As shown, the calibration curve evaluated the consistency between the nomogram prediction and the actual observation results, indicating that there is good consistency between the nomogram prediction and the actual observation, and the effect of predicting survival rate by the nomogram is quite considerable.

[0093] The experimental results of the above embodiments show that the risk score based on 14 LRGs can well distinguish hepatocellular carcinoma patients into high-risk and low-risk groups, which may provide an effective guide for the prognosis management of hepatocellular carcinoma patients.

[0094] The above embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work should fall within the protection scope of the present invention.

Claims

1. A method for constructing a lysosome-related gene prognostic model for hepatocellular carcinoma, characterized in that: The following steps are involved: Step 1) Download the hepatocellular carcinoma-related training set from the Cancer Genome Atlas database, which contains standardized RNA-seq data sets and clinical data, and select hepatocellular carcinoma patients with a survival time of more than 90 days and survival status and clinical information as the training set; download the standardized RNA-seq data set and clinical data of hepatocellular carcinoma samples from the International Cancer Genome Consortium database as the validation set; Step 2) Obtain lysosome-related genes from the molecular feature database and use Cox regression analysis to perform prognostic analysis on lysosome-related genes. P <0.05 was used as the cutoff value for screening prognosis-related genes; Step 3) Using the initial candidate prognosis-related lysosomal genes, LASSO regression analysis was performed in the training set to screen out genes for constructing a risk score model; The risk score is expressed as: ； in, n is the number of prognostic genes, exp i For gene i The expression value of β i For gene i The regression coefficient of The lysosome-related genes include: ATG10, ATP6V0B, HMOX1, RAB7A, SQSTM1, TMEM41B, TP53, FTH1, RNF19B, RAMP3, TPP1, AP1M2, CLTA, DNAJC6; The risk scoring model is as follows: Risk score = (0.1771 × ATG10 gene expression level) + (0.2816 × ATP6V0B gene expression level) + (0.0735 × HMOX1 gene expression level) + (0.198 × RAB7A gene expression level) + (0.0305 × SQSTM1 gene expression level) + (0.1693 × TMEM41B gene expression level) + (-0.0066 × TP53 gene expression level) + (-0.0746 × FTH1 gene expression level) + (0.037 × RNF19B gene expression level) + (0.1362 × RAMP3 gene expression level) + (0.0504 × TPP1 gene expression level) + (0.0122 × AP1M2 gene expression level) + (0.1078 × CLTA gene expression level) + (0.0062 × DNAJC62 gene expression level); Step 4) Use the coefficient of each gene to calculate an individualized risk score, and divide HCC patients into high-risk and low-risk groups using the median risk score; use ROC to calculate the area under the curve (AUC) at multiple time points to evaluate the discriminatory ability of the prognostic model; finally, use the same risk score formula and critical value in the validation set to verify the accuracy of the model.

2. The method for constructing a lysosome-related gene prognostic model for hepatocellular carcinoma according to claim 1, characterized in that: The "ATG10" refers to a nucleic acid encoding all or part of the ATG10 protein or substantially identical to all or part of the nucleic acid sequence or its analog, and its Gene ID is 83734; The "ATP6V0B" refers to a nucleic acid encoding all or part of the ATP6V0B protein or being substantially identical to all or part of the nucleic acid sequence or its analogs, and its Gene ID is 533; The "HMOX1" refers to a nucleic acid encoding all or part of the HMOX1 protein or substantially identical to all or part of the nucleic acid sequence or its analog, and its Gene ID is 3162; The "RAB7A" refers to a nucleic acid encoding all or part of the RAB7A protein or substantially identical to all or part of the nucleic acid sequence or its analogs, and its Gene ID is 7879; The "SQSTM1" refers to a nucleic acid encoding all or part of the SQSTM1 protein or substantially identical to all or part of the nucleic acid sequence or its analog, and its Gene ID is 8878; The "TMEM41B" refers to a nucleic acid encoding all or part of the TMEM41B protein or substantially identical to all or part of the nucleic acid sequence or its analog, and its Gene ID is 440026; The "TP53" refers to a nucleic acid encoding all or part of the TP53 protein or substantially identical to all or part of the nucleic acid sequence or its analog, and its Gene ID is 7157; The "FTH1" refers to a nucleic acid encoding all or part of the FTH1 protein or substantially identical to all or part of the nucleic acid sequence or its analog, and its Gene ID is 2495; The "RNF19B" refers to a nucleic acid encoding all or part of the RNF19B protein or substantially identical to all or part of the nucleic acid sequence or its analog, and its Gene ID is 127544; The "RAMP3" refers to a nucleic acid encoding all or part of the RAMP3 protein or substantially identical to all or part of the nucleic acid sequence or its analogs, and its Gene ID is 10268; The "TPP1" refers to a nucleic acid encoding all or part of the TPP1 protein or substantially identical to all or part of the nucleic acid sequence or its analog, and its Gene ID is 1200; The "AP1M2" refers to a nucleic acid encoding all or part of the AP1M2 protein or substantially identical to all or part of the nucleic acid sequence or its analogs, and its Gene ID is 10053; The "CLTA" refers to a nucleic acid encoding all or part of the CLTA protein or substantially identical to all or part of the nucleic acid sequence or its analog, and its Gene ID is 1211; The “DNAJC6” refers to a nucleic acid encoding all or part of the DNAJC6 protein or being approximately identical to all or part of the nucleic acid sequence or its analogs, and its Gene ID is 9829.

3. Use of the prognostic model constructed by the construction method according to any one of claims 1 or 2 in preparing a product for predicting the prognosis of hepatocellular carcinoma, characterized in that: The sample for detecting the expression of characteristic genes is tumor tissue.

4. The use according to claim 3, characterized in that: The product is a kit or a chip.

5. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: The processor performs the following steps: Step A) obtaining the expression data of lysosome-related genes of the hepatocellular carcinoma patient to be tested; Step B) inputting the expression data of lysosome-related genes of the hepatocellular carcinoma patient to be tested into the prognostic model constructed by the construction method according to claim 1 or 2, and calculating the risk score of the hepatocellular carcinoma patient to be tested; Step C) predicting the overall survival rate of patients with hepatocellular carcinoma based on the risk score.

6. A computer-readable storage medium having a computer program stored thereon, characterized in that: The computer program is executed by the processor to perform the following steps: Step A) obtaining the expression data of lysosome-related genes of the hepatocellular carcinoma patient to be tested; Step B) inputting the expression data of lysosome-related genes of the hepatocellular carcinoma patient to be tested into the prognostic model constructed by the construction method according to claim 1 or 2, and calculating the risk score of the hepatocellular carcinoma patient to be tested; Step C) predicting the overall survival rate of patients with hepatocellular carcinoma based on the risk score.

7. A computer program product, comprising a computer program, characterized in that The computer program is executed by the processor to perform the following steps: Step A) obtaining the expression data of lysosome-related genes of the hepatocellular carcinoma patient to be tested; Step B) inputting the expression data of lysosome-related genes of the hepatocellular carcinoma patient to be tested into the prognostic model constructed by the construction method according to claim 1 or 2, and calculating the risk score of the hepatocellular carcinoma patient to be tested; Step C) predicting the overall survival rate of patients with hepatocellular carcinoma based on the risk score.

Images