Prognosis evaluation method and system for diffuse large B-cell lymphoma

Abstract

The invention discloses a prognosis evaluation method and system for diffuse large B-cell lymphoma. According to the method, firstly, a gene module closely related to lipid metabolism is screened from DLBCL transcriptome data through weighted gene co-expression network analysis (WGCNA), and then eight key prognosis genes including FNDC1, IL22RA2, C15orf48, OMD, MFAP2, BC017398, CXCL6 and TNFAIP6 are identified from the module by adopting multi-step regression analysis (single factor Cox, LASSO and multi-factor Cox). A risk scoring model is constructed based on the expression levels and regression coefficients of the genes, and DLBCL patients can be divided into a high-risk group and a low-risk group with significant survival differences. The risk score and the clinical pathological factors are further integrated to construct a column graph, and individualized survival probability prediction can be achieved. The invention further provides a corresponding prognosis evaluation system, electronic equipment and a storage medium. An independent data set verifies that the prognosis model has excellent prediction performance and clinical practical value.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of bioinformatics and medical prognosis evaluation, and in particular to a diffuse large B-cell lymphoma prognosis evaluation method and system. BACKGROUND

[0002] Diffuse large B-cell lymphoma (DLBCL) is the most common aggressive non-Hodgkin lymphoma, accounting for about 40% of adult cases. Although the first-line R-CHOP treatment regimen and new therapies such as CAR-T have improved patient prognosis, about one-third of patients still develop drug resistance or recurrence, and some patients respond poorly to CAR-T therapy, and the underlying mechanisms have not been fully elucidated. Therefore, developing a tool that can accurately evaluate the prognosis of DLBCL patients and identify high-risk groups is of great significance for achieving individualized treatment.

[0003] Lipid metabolism plays a key role in tumor development and treatment response. Studies have shown that lipid metabolism disorder is one of the most prominent metabolic changes in cancer, and cancer cells support their proliferation, survival and metastasis by regulating lipid metabolism, and affect the tumor microenvironment. In DLBCL, lipid metabolism characteristics are associated with treatment resistance and poor prognosis, for example, the content of specific lipid fragments is increased in drug-resistant samples, and lipid metabolism disorder M2 macrophages can induce T cell function inhibition through cholesterol efflux. These findings suggest that lipid metabolism plays an important role in DLBCL prognosis and treatment response, but the specific regulatory mechanisms remain to be further explored.

[0004] Currently, some DLBCL prognosis models based on gene expression characteristics have been developed, such as models based on other gene characteristics. However, most of the existing models fail to fully utilize lipid metabolism gene information closely related to the development of DLBCL, or the robustness and prediction accuracy of the constructed model need to be further improved. In addition, many models fail to provide an individualized prognosis prediction tool that can integrate key prognostic factors and directly serve clinical practice, thereby limiting their practical application value in precision medicine.

[0005] Therefore, there is an urgent need in the art for a new technical solution that can fully utilize lipid metabolism-related gene information, construct an accurate and reliable prognosis evaluation model, and provide individualized prognosis prediction for clinical decision-making. SUMMARY

[0006] In view of the above problems, the purpose of the present application is to provide a diffuse large B-cell lymphoma prognosis evaluation method and system.

[0007] In order to achieve the above technical purpose, the technical scheme adopted by the present application is as follows: In a first aspect, the present application provides a diffuse large B-cell lymphoma prognosis evaluation method, comprising the following steps: S1: obtaining gene expression data of a DLBCL patient to be evaluated; S2: inputting the gene expression data into a prognosis risk score model, which is constructed based on the expression levels of eight genes FNDC1, IL22RA2, C15orf48, OMD, MFAP2, BC017398, CXCL6 and TNFAIP6 and their respective regression coefficients; S3: dividing the patient into different prognosis risk groups based on the risk score output by the model.

[0008] Further, the risk score S output by the model is calculated according to the following formula:

[0009] wherein, represents the gene expression level of FNDC1, represents the regression coefficient corresponding to FNDC1, = -1.496; represents the gene expression level of IL22RA2, represents the regression coefficient corresponding to IL22RA2, = -0.559; represents the gene expression level of C15orf48, represents the regression coefficient corresponding to C15orf48, = -1.561; represents the gene expression level of OMD, represents the regression coefficient corresponding to OMD, = -1.053; represents the gene expression level of MFAP2, represents the regression coefficient corresponding to MFAP2, = 2.873; represents the gene expression level of BC017398, represents the regression coefficient corresponding to BC017398, = 0.973; represents the gene expression level of CXCL6, represents the regression coefficient corresponding to CXCL6, = 0.771; represents the gene expression level of TNFAIP6, represents the regression coefficient corresponding to TNFAIP6, = 1.451.

[0010] Further, the method for prognosis evaluation of diffuse large B-cell lymphoma further comprises S4: constructing nomogram for predicting individual survival probability based on the risk score and in combination with at least one clinicopathological factor.

[0011] Further, the clinicopathological factor comprises age, gender and tumor stage.

[0012] Further, the method for constructing the prognosis risk score model comprises: Based on the DLBCL transcriptome data, a gene module related to lipid metabolism phenotype is identified by weighted gene co-expression network analysis; From the gene module, the eight key genes are screened and the regression coefficients are determined by a combination of multiple regression analyses. The combination of multiple regression analyses comprises sequentially performing single factor Cox regression analysis, LASSO regression analysis and multi-factor Cox regression analysis.

[0013] Further, the method for constructing the prognosis risk score model specifically comprises the following steps: 1) Data acquisition and pretreatment: The training set and the validation set are obtained from the GEO database; all gene expression data are subjected to standardization processing, including background correction, quantile normalization and probe annotation, to eliminate batch effects and technical differences; 2) Key gene module identification: The training set is subjected to in-depth analysis by weighted gene co-expression network analysis, and the specific steps are as follows: Data pretreatment: calculate the expression coefficient of variation of all genes, and select the top 5000 genes with the highest coefficient of variation for network construction; Network construction: construct a gene similarity matrix by Pearson correlation coefficient, and use the pickSoftThreshold function to evaluate and determine the optimal soft threshold β=2, at this time the scale topological fitting index R 2 >0.85, which meets the scale-free network characteristics; Module identification: hierarchical clustering is performed based on the topological overlap matrix, and the dynamic pruning tree algorithm is used to divide the initial module; the minimum module gene number is set to 30, and the merging threshold is set to 0.25 (i.e. the modules with a correlation coefficient between the module characteristic genes greater than 0.75 are merged), and finally a plurality of characteristic modules are obtained; Module-phenotype association: calculate the correlation between each module characteristic gene and lipid metabolism phenotype, and verify the correlation by module member and gene characteristic correlation analysis, thereby determining the key module; 3) Construction of prognosis risk model: Based on the key module genes, a prognosis risk model is constructed by multi-step regression analysis, specifically as follows: The genes significantly related to prognosis were screened by single factor Cox regression; 10-fold cross-validation was used to further compress variables to obtain key prognostic factors; 8 independent prognostic genes were finally determined by multivariate Cox regression: FNDC1, IL22RA2, C15orf48, OMD, MFAP2, BC017398, CXCL6 and TNFAIP6; A prognostic risk score model was constructed based on the expression levels of the above 8 genes and their respective regression coefficients.

[0014] Further, in step S3, the patients are divided into a high-risk group and a low-risk group according to the median of the risk score.

[0015] In a second aspect, the present application provides a system for evaluating the prognosis of diffuse large B-cell lymphoma, which applies the prognostic evaluation method as described in the first aspect above, and the system comprises: A data acquisition module for acquiring or receiving gene expression data of a sample to be tested; A storage module pre-stored with a prognostic risk score model constructed based on the expression levels of the eight genes FNDC1, IL22RA2, C15orf48, OMD, MFAP2, BC017398, CXCL6 and TNFAIP6 and their regression coefficients; A processing module configured to call the prognostic risk score model and process the gene expression data to calculate a risk score; An output module for outputting the risk score and the prognosis risk evaluation result based on the score.

[0016] In a third aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the prognostic evaluation method as described in the first aspect above.

[0017] In a fourth aspect, an electronic device comprises a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the prognostic evaluation method as described in the first aspect above when executing the program.

[0018] In a fifth aspect, the present application provides a kit for evaluating the prognosis of diffuse large B-cell lymphoma; the kit comprises detection reagents for detecting the expression of the eight lipid metabolism-related genes (FNDC1, IL22RA2, C15orf48, OMD, MFAP2, BC017398, CXCL6 and TNFAIP6). The detection reagents can include specific primers, probes or gene chips for the eight genes, etc.

[0019] Compared with the prior art, the application has the beneficial effects that: 1. Novel markers and mechanism-related: The application first screened and verified a marker set composed of eight lipid metabolism-related genes (FNDC1, IL22RA2, C15orf48, OMD, MFAP2, BC017398, CXCL6, and TNFAIP6) for DLBCL prognostic evaluation. These genes are closely related to lipid metabolism and immune microenvironment regulation, providing more reliable biological mechanism for prognosis prediction.

[0020] 2. Scientific model construction method: The application uses a systematic strategy of WGCNA combined with various regression analyses (single-factor Cox, LASSO, and multi-factor Cox) to construct a risk score model, effectively screens key prognostic factors, avoids overfitting, and ensures the stability and generalization ability of the model. The model is effectively verified in an independent validation set.

[0021] 3. Excellent prediction performance: The eight-gene-based risk score model provided by the application exhibits good prognostic discrimination ability in both the training set and the validation set, and the overall survival time of the high-risk group is significantly shorter than that of the low-risk group. Time-dependent ROC curve analysis shows that the model has high prediction accuracy (AUC value) at 1 year, 3 years, and 5 years.

[0022] 4. High clinical practical value: The risk score is confirmed to be an independent prognostic factor independent of age, gender, and stage, and has independent prognostic value. The nomogram constructed by further integrating the risk score and clinical factors provides an intuitive individualized survival probability prediction tool, and the predicted value is highly consistent with the actual observed value (calibration curve verification). Decision curve analysis (DCA) shows that it can provide the maximum net benefit in clinical decision-making. BRIEF DESCRIPTION OF DRAWINGS

[0023] In order to more clearly illustrate the technical solutions in the embodiments of the application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiment or prior art description. Obviously, the drawings in the following description only constitute some embodiments of the application, and for those skilled in the art, other drawings can also be obtained without creative labor on the basis of these drawings.

[0024] Figure 1 Flowchart for identifying key gene modules by WGCNA.

[0025] Figure 2 Soft threshold parameter screening diagram showing that the network meets the scale-free distribution characteristics when power=2.

[0026] Figure 3Hierarchical clustering tree map for gene co-expression modules.

[0027] Figure 4 Heat map for correlation of module-specific genes with lipid metabolism phenotypes.

[0028] Figure 5 Scatter plot for module membership of genes in the blue module versus gene features.

[0029] Figure 6 LASSO regression coefficient path plot.

[0030] Figure 7 LASSO regression 10-fold cross-validation error curve plot.

[0031] Figure 8 Forest plot of eight key prognostic genes and their regression coefficients determined by multivariate Cox regression analysis.

[0032] Figure 9 Heat map showing expression of eight key prognostic genes in validation set samples.

[0033] Figure 10 Decision curve analysis (DCA) results showing nomograms. DETAILED DESCRIPTION

[0034] The application will be further described below in conjunction with the accompanying drawings and embodiments. It is particularly pointed out that the following embodiments are only used to illustrate the application, but do not limit the scope of the application. Similarly, the following embodiments are only part of the embodiments of the application, not all embodiments, and all other embodiments obtained by those of ordinary skill in the art without creative labor are within the scope of the application.

[0035] The application provides a prognostic evaluation method for diffuse large B-cell lymphoma (DLBCL), which is based on a marker set of eight specific lipid metabolism-related genes. The method mainly includes the following steps: S1: obtaining gene expression data of a DLBCL patient to be evaluated; S2: inputting the gene expression data into a prognostic risk score model, which is constructed based on the expression levels of eight genes FNDC1, IL22RA2, C15orf48, OMD, MFAP2, BC017398, CXCL6 and TNFAIP6 and their respective regression coefficients; S3: dividing the patient into different prognostic risk groups, such as a high-risk group and a low-risk group, based on the risk score output by the model, thereby achieving prognostic risk evaluation. The risk score model is constructed based on the expression levels of the above eight genes and the regression coefficients, and the calculation formula is: Risk score =∑(gene expression level x corresponding regression coefficient) Next, the patients are divided into a high-risk group and a low-risk group according to the median of the risk score.

[0036] S4: Based on the risk score, and in combination with at least one clinicopathological factor (preferably age, gender, and tumor stage), a nomogram is constructed for predicting individualized survival probability (such as 1-year, 3-year, and 5-year survival rate).

[0037] In the present application, the construction of the prognostic risk score model employs a systematic bioinformatics procedure: 1) First, based on the DLBCL transcriptome data, a gene module (such as the blue module) significantly associated with the lipid metabolism phenotype is identified through weighted gene co-expression network analysis (WGCNA). The WGCNA analysis process includes data preprocessing and high-variant gene screening, construction of a co-expression network based on Pearson correlation coefficient and a selected soft threshold (such as β = 2), hierarchical clustering and dynamic pruning based on topological overlap matrix (TOM) to divide and merge modules (for example, the merging threshold is set to 0.25), and locking the key module through the association analysis of the module characteristic gene and the lipid metabolism phenotype.

[0038] 2) From the key module, eight key genes are finally selected through a combination of multiple regression analysis strategies, and their regression coefficients are determined. The selection strategy preferably includes the following in order: single-factor Cox regression analysis for preliminary screening, LASSO regression analysis (for example, using 10-fold cross-validation) for variable compression and preventing overfitting, and multi-factor Cox regression analysis to determine independent prognostic genes and their weights (regression coefficients).

[0039] The present application will be further described in detail below in conjunction with the accompanying drawings and examples: Example 1 Referring to the accompanying drawings, Figure 1 the present embodiment provides a prognostic evaluation method for diffuse large B-cell lymphoma, which comprises: S1: Obtain gene expression data of a DLBCL patient to be evaluated; S2: Input the gene expression data into a prognostic risk score model, which is constructed based on the expression levels of the eight genes FNDC1, IL22RA2, C15orf48, OMD, MFAP2, BC017398, CXCL6, and TNFAIP6 and their respective corresponding regression coefficients; In the present embodiment, the prognostic risk score model is constructed in the following manner: 1) Data acquisition and preprocessing The training set and validation set were obtained from GEO database. All gene expression data were standardized, including background correction, quantile normalization and probe annotation, to eliminate batch effects and technical differences.

[0040] 2) Key gene module identification Weighted gene co-expression network analysis (WGCNA) was used to analyze the training set in depth; the specific steps are as follows: Data preprocessing: Calculate the expression coefficient of variation of all genes, and select the top 5000 genes with the highest coefficient of variation for network construction; Network construction: Construct gene similarity matrix through Pearson correlation coefficient, and use pickSoftThreshold function to evaluate and determine the optimal soft threshold β = 2 (such as Figure 2 ), At this time, the scale topological fitting index R²> 0.85, which meets the scale-free network characteristics; Module identification: Hierarchical clustering based on topological overlap matrix (TOM), using dynamic pruning tree algorithm to divide the initial module, setting the minimum module gene number to 30, and the merging threshold to 0.25 (i.e. The modules with correlation coefficient greater than 0.75 between module characteristic genes are merged), and finally 5 characteristic modules are obtained: brown (307 genes), blue (384 genes), yellow (91 genes), green (58 genes) and turquoise (4160 genes) (such as Figure 3 ); Module-phenotype correlation: Calculate the correlation between each module characteristic gene and lipid metabolism phenotype, and find that the blue module is significantly positively correlated with lipid metabolism characteristics (cor = 0.54, p = 1e-32) (such as Figure 4 ), Module member and gene characteristic correlation analysis further verifies the correlation (such as Figure 5 ).

[0041] 3) Prognostic risk model construction and verification Based on the blue module genes, a multi-step regression analysis was used to construct a prognostic risk model: Univariate Cox regression: 172 genes significantly associated with prognosis (p<0.05) were selected; LASSO regression: 10-fold cross-validation was used to further compress variables, and 42 key prognostic factors were obtained (such as Figures 6-7 ); Multivariate Cox regression: Finally, 8 independent prognostic genes were determined: FNDC1, IL22RA2, C15orf48, OMD, MFAP2, BC017398, CXCL6 and TNFAIP6 (such as Figure 8), wherein CXCL6, BC017398, TNFAIP6, MFAP2 are risk factors, C15orf48, FNDC1, OMD, IL22RA2 are protective factors. The regression coefficients of each gene are as follows Figure 8 .

[0042] Based on the above gene expression levels and regression coefficients, a prognosis risk score model was constructed, and the calculation formula was:

[0043] wherein, represents the gene expression level of FNDC1, represents the regression coefficient corresponding to FNDC1, = -1.496; represents the gene expression level of IL22RA2, represents the regression coefficient corresponding to IL22RA2, = -0.559; represents the gene expression level of C15orf48, represents the regression coefficient corresponding to C15orf48, = -1.561; represents the gene expression level of OMD, represents the regression coefficient corresponding to OMD, = -1.053; represents the gene expression level of MFAP2, represents the regression coefficient corresponding to MFAP2, = 2.873; represents the gene expression level of BC017398, represents the regression coefficient corresponding to BC017398, = 0.973; represents the gene expression level of CXCL6, represents the regression coefficient corresponding to CXCL6, = 0.771; represents the gene expression level of TNFAIP6, represents the regression coefficient corresponding to TNFAIP6, = 1.451.

[0044] According to the median of the risk score, the patients were divided into a high-risk group and a low-risk group. In the training set, the overall survival time of the patients in the high-risk group was significantly shorter than that in the low-risk group. The time-dependent ROC curve showed that the model had high prediction accuracy (AUC value) at 1 year, 3 years, and 5 years. The risk score distribution, survival status distribution, and gene expression heat map (such as Figure 9 ) further verified the model performance.

[0045] The above analysis was repeated in an independent validation set, and the results showed that the model maintained good prediction performance, confirming its good generalization ability.

[0046] 4) Clinical validation and nomogram construction To evaluate the clinical value of the risk score, the following analysis was performed: Independent validation: Univariate and multivariate Cox regression analysis showed that the risk score was an independent prognostic factor independent of age, sex, and stage; Distribution of clinical characteristics: There was no significant difference in age, sex, stage, and other basic clinical characteristics between the high-risk group and the low-risk group, indicating that the risk stratification was independent; Nomogram construction: The risk score, age, sex, and stage were integrated to construct a prognostic prediction nomogram for individualized prediction of 1-year, 3-year, and 5-year survival probability.

[0047] The nomogram verification results showed that: The calibration curve showed that the predicted survival rate was highly consistent with the actual observed value; the ROC curve showed that the prediction performance of the nomogram was significantly better than that of a single variable; the decision curve analysis showed that the nomogram obtained the maximum net benefit at all time points (e.g. Figure 10 ).

[0048] S3: Based on the risk score output by the model, patients are divided into different prognostic risk groups, such as high-risk group and low-risk group, to achieve prognostic risk assessment.

[0049] S4: Based on the risk score, and combined with at least one clinical pathological factor (preferably age, sex, and tumor stage), a nomogram for predicting individualized survival probability (such as 1-year, 3-year, and 5-year survival rate) is constructed.

[0050] Example 2 The present embodiment provides a system for prognostic evaluation of diffuse large B-cell lymphoma, which applies the prognostic evaluation method of diffuse large B-cell lymphoma in the above-mentioned Example 1, and the system comprises: A data acquisition module for acquiring or receiving gene expression data of a sample to be tested; A storage module pre-stored with a prognostic risk score model constructed based on the expression levels of the eight genes FNDC1, IL22RA2, C15orf48, OMD, MFAP2, BC017398, CXCL6, and TNFAIP6 and their regression coefficients (such as the prognostic risk score model constructed in Example 1); A processing module configured to call the prognostic risk score model and process the gene expression data to calculate the risk score; An output module is configured to output the risk score and a prognosis risk assessment result based on the risk score.

[0051] Embodiment 3 Based on the same inventive concept, the embodiment provides a computer readable storage medium storing computer instructions, which, when executed on a computer, cause the computer to perform the prognosis evaluation method for diffuse large B-cell lymphoma as in Embodiment 1.

[0052] In the specific implementation process, the computer readable storage medium includes: a Universal Serial Bus flash drive (USB), a mobile hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and various storage media that can store program codes.

[0053] The apparatus embodiments described above are merely illustrative, and units / modules illustrated as separate components can or can not be physically separated, and the components illustrated as units / modules can or can not be physical units / modules, i.e., can be located in one place or distributed on multiple network units / modules. Part or all of the modules can be selected to achieve the purposes of the embodiments according to actual needs. Those skilled in the art can understand and implement it without creative labor.

[0054] From the above description of the embodiments, those skilled in the art can clearly understand that the embodiments can be implemented by means of software and necessary universal hardware platforms, and of course can also be implemented by hardware. Based on such understanding, the above technical solutions, essentially or in other words, the part that contributes to the prior art, can be embodied in the form of a software product, which can be stored in a computer readable storage medium, such as a ROM / RAM, a magnetic disk, an optical disk, etc., and includes a number of instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute the methods of the embodiments or some parts of the embodiments.

[0055] Embodiment 4 The embodiment provides an electronic device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the prognosis evaluation method for diffuse large B-cell lymphoma as in Embodiment 1 when executing the program.

[0056] Embodiment 5 This embodiment describes the preparation of a kit for prognostic assessment of DLBCL based on the eight key genes mentioned above (FNDC1, IL22RA2, C15orf48, OMD, MFAP2, BC017398, CXCL6, and TNFAIP6). The kit contains: Specific primers and probes for the FNDC1, IL22RA2, C15orf48, OMD, MFAP2, BC017398, CXCL6 and TNFAIP6 genes; Various reagents required for RT-PCR reactions; Positive and negative control samples; The instruction manual provides a detailed description of the testing process and risk score calculation method.

[0057] When using this method, RNA is extracted from the sample to be tested, reverse transcribed into cDNA, and the expression levels of eight genes are detected by quantitative PCR. The expression levels are then substituted into the risk scoring formula to calculate the risk score, and the patient's prognostic risk is finally assessed based on the score results.

[0058] The above description is only a part of the embodiments of the present invention and does not limit the scope of protection of the present invention. Any equivalent device or equivalent process transformation made based on the content of the present invention specification and drawings, or direct or indirect application in other related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. A method for prognostic evaluation of diffuse large B-cell lymphoma, characterized by, Comprising the following steps: S1: obtaining gene expression data of a DLBCL patient to be evaluated; S2: inputting the gene expression data into a prognostic risk score model, which is constructed based on the expression levels of eight key genes FNDC1, IL22RA2, C15orf48, OMD, MFAP2, BC017398, CXCL6 and TNFAIP6 and their respective regression coefficients; S3: dividing the patient into different prognostic risk groups based on the risk score output by the model.

2. The method for evaluating the prognosis of diffuse large B-cell lymphoma according to claim 1, wherein The risk score S output by the model is calculated as follows: wherein, represents a gene expression level of FNDC1, represents a regression coefficient corresponding to FNDC1, = -1.496; represents a gene expression level of IL22RA2, represents a regression coefficient corresponding to IL22RA2, = -0.559; represents a gene expression level of C15orf48, represents a regression coefficient corresponding to C15orf48, = -1.561; represents a gene expression level of OMD, represents a regression coefficient corresponding to OMD, = -1.053; represents a gene expression level of MFAP2, represents a regression coefficient corresponding to MFAP2, = 2.873; represents a gene expression level of BC017398, represents a regression coefficient corresponding to BC017398, = 0.973; represents a gene expression level of CXCL6, represents a regression coefficient corresponding to CXCL6, = 0.771; represents a gene expression level of TNFAIP6, represents a regression coefficient corresponding to TNFAIP6, = 1.

451.

3. The method for prognosis evaluation of diffuse large B-cell lymphoma according to Claim 1, wherein Also comprising: S4: based on the risk score and in combination with at least one clinicopathological factor, constructing a nomogram for predicting individual survival probability.

4. The method for prognosis evaluation of diffuse large B-cell lymphoma according to claim 3, characterized in that, The clinicopathological factors include age, gender and tumor stage.

5. The method of claim 1, wherein the diffuse large B-cell lymphoma is characterized by, The construction method of the prognostic risk score model comprises: Based on the DLBCL transcriptome data, identify the gene module related to lipid metabolism phenotype through weighted gene co-expression network analysis; From the gene module, screen the eight key genes and determine their regression coefficients through a combination of multiple regression analyses; The combination of multiple regression analyses includes sequentially performing single factor Cox regression analysis, LASSO regression analysis and multi-factor Cox regression analysis.

6. The method for the prognosis evaluation of diffuse large B-cell lymphoma according to claim 5, characterized in that, The construction method of the prognostic risk score model specifically comprises the following steps: 1) Data acquisition and preprocessing: Obtain the training set and validation set from the GEO database; all gene expression data are standardized, including background correction, quantile normalization and probe annotation, to eliminate batch effects and technical differences; 2) Key gene module identification: In-depth analysis of the training set using weighted gene co-expression network analysis, the specific steps are as follows: Data preprocessing: calculate the expression coefficient of variation of all genes, and select the top 5000 genes with the highest coefficient of variation for network construction; Network construction: The gene similarity matrix was constructed by Pearson correlation coefficient, and the optimal soft threshold β = 2 was determined by using pickSoftThreshold function to evaluate. At this time, the scale topological fitting index R 2 >0.85, which meets the characteristics of scale-free network; Module identification: hierarchical clustering based on topological overlap matrix, using dynamic pruning tree algorithm to divide the initial module, setting the minimum module gene number to 30 and the merging threshold to 0.25, finally obtaining several characteristic modules; Module-phenotype association: calculate the correlation between each module characteristic gene and lipid metabolism phenotype, and verify the correlation through module member and gene characteristic correlation analysis, to determine the key module; 3) Prognostic risk model construction: Based on the key module genes, use multi-step regression analysis to construct the prognostic risk model, as follows: Through single factor Cox regression, screen out genes significantly related to prognosis; use 10-fold cross-validation to further compress variables, obtain key prognostic factors; through multi-factor Cox regression, finally determine eight independent prognostic genes: FNDC1, IL22RA2, C15orf48, OMD, MFAP2, BC017398, CXCL6 and TNFAIP6; Based on the expression levels of the above eight genes and their respective regression coefficients, construct a prognostic risk score model.

7. The method of claim 1, wherein the diffuse large B-cell lymphoma is prognostic evaluation method, characterized in that, In step S3, the patients are divided into high-risk group and low-risk group according to the median of the risk score.

8. A system for diffuse large B-cell lymphoma prognosis evaluation, characterized in that, The application discloses a prognosis evaluation method and a system. A data acquisition module is configured to acquire or receive gene expression data of a sample to be tested. A storage module pre-stores a prognosis risk score model based on expression levels of eight genes, i.e., FNDC1, IL22RA2, C15orf48, OMD, MFAP2, BC017398, CXCL6 and TNFAIP6, and regression coefficients thereof. A processing module is configured to call the prognosis risk score model and process the gene expression data to calculate a risk score. An output module is configured to output the risk score and a prognosis risk evaluation result based on the score.

9. A computer readable storage medium having stored thereon a computer program, characterized in that, The program is executed by a processor to implement the method of any one of claims 1-7.

10. An electronic device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, The processor executes the program to implement the method of any one of claims 1-7.

Images