Colorectal cancer metabolism-related prognosis model construction method based on multiple omics

By constructing a multi-omics-based metabolism-related prognostic model for colorectal cancer, the problem of lack of metabolic gene research in existing technologies has been solved, more accurate prognosis prediction and personalized treatment have been achieved, and the survival rate and treatment effect of colorectal cancer patients have been improved.

CN120708862APending Publication Date: 2025-09-26YANTAI AFFILIATED HOSPITAL OF BINZHOU MEDICAL COLLEGE
View PDF 0 Cites 1 Cited by

Patent Information

Application Number
CN202510366900.6
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-03-26
Publication Date
2025-09-26

AI Technical Summary

Technical Problem

The existing prognostic model for colorectal cancer lacks research based on metabolic genes, resulting in poor treatment effects for patients in the middle and late stages, low 5-year survival rates, and difficulty in early diagnosis and treatment.

Method used

A multi-omics-based metabolic-related prognostic model for colorectal cancer was constructed. RNASeq data were obtained from the TCGA database, differential expression analysis was performed, and prognosis-related metabolic genes were screened. The model was combined with nomogram and survival analysis, and multiple R software packages were used for model validation and evaluation, integrating multi-omics data for prediction.

Benefits of technology

It improves the accuracy and personalization of prognosis prediction for colorectal cancer patients, provides more scientific and precise treatment plans, reduces the risk of recurrence and metastasis, and is applicable to prognosis prediction and personalized treatment of other types of cancer.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN120708862A_ABST
    Figure CN120708862A_ABST
Patent Text Reader

Abstract

The invention discloses a colorectal cancer metabolism-related prognosis model construction method based on multiple omics. By integrating multiple omics data, the defect is overcome, a new view angle and method are provided, the accuracy and practicability of the prognosis model are effectively improved, and the blank in the prior art is filled. Meanwhile, the method provided by the invention is not only suitable for prognosis prediction of colorectal cancer patients, but also can be popularized and applied to other types of cancers. As the metabolism-related gene plays an important role in various cancers, the technical scheme provided by the invention has relatively high universality and popularization value, and can provide reference for prognosis prediction and personalized treatment of other cancers. Finally, the prognosis model provided by the invention can be used as a clinical decision support tool to help doctors to make more scientific and accurate decisions in diagnosis and treatment processes. Through early recognition of high-risk patients and implementation of targeted intervention measures, prognosis of the patients can be significantly improved, and the risk of relapse and metastasis can be reduced.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the technical field of colorectal cancer model analysis, and specifically is a method for constructing a colorectal cancer metabolism-related prognosis model based on multi-omics. Background Art

[0002] Colorectal cancer (CRC) is a common malignancy. Its development and progression are the result of a synergistic interaction of multiple factors, including genetics, lifestyle, obesity, and the environment, and involve changes at various molecular levels, including the genome, transcriptome, methylation, and epigenome. The main pathways involved include: ① CIN pathway; ② MSI pathway; ③ CIMP pathway; and ④ serrated pathway. Although screening, diagnosis, and treatment for CRC have greatly improved compared to previous generations, and early diagnosis and treatment have significantly improved overall survival for patients in the early stages, treatment efficacy remains poor for patients in the middle and late stages, and the prognosis remains relatively poor. The 5-year survival rate is approximately 15%. Abnormal tumor metabolism is a major factor affecting patient prognosis. Metabolic reprogramming is a hallmark of cancer, closely associated with tumorigenesis and immune evasion. Metabolic remodeling can influence cancer cell proliferation, differentiation, and therapeutic epigenetic changes. Therefore, identifying new diagnostic and therapeutic strategies based on tumor metabolism to improve the prognosis of CRC patients is an urgent clinical challenge.

[0003] Early-stage CRC patients can be eradicated using endoscopic and surgical treatment. However, because most CRC patients lack obvious clinical symptoms in the early stages, approximately 60% are diagnosed at advanced stages, with lymph node and distant metastases. The 5-year survival rate for stage IV patients drops to approximately 15%. This suggests that early diagnosis and treatment of CRC can yield a favorable prognosis. Therefore, developing effective early prognosis prediction models is crucial.

[0004] However, despite increasing evidence that gene-based prognostic models play an important role in various human cancers, prognostic models based on metabolic genes in colorectal cancer are still lacking, especially those based on big data research methods. Summary of the Invention

[0005] The purpose of the present invention is to provide a method for constructing a colorectal cancer metabolism-related prognosis model based on multi-omics in order to solve the above-mentioned problems.

[0006] The technical solution adopted by the present invention is as follows: a method for constructing a colorectal cancer metabolism-related prognostic model based on multi-omics, the method comprising the following steps:

[0007] S1: RNASeq data of TCGA-CRC were obtained from the TCGA database;

[0008] S2: Differential expression analysis was performed: Wilcox test and R package “edgeR” were used to screen differentially expressed metabolic genes in CRC;

[0009] S3: Screening of differentially expressed genes related to prognosis;

[0010] S4: Construction of prognosis-related metabolic gene model and survival analysis in colorectal cancer;

[0011] S5: Perform nomogram analysis and consistency index: Use the nomogram combined with clinicopathological characteristics (TNM stage, pathological stage, age, sex) and risk score to predict the 1-, 3-, and 5-year overall survival of CRC patients;

[0012] S6: Decision curve analysis was performed: DCA curves were drawn using the R package “ggDCA”. Clinically relevant ROC curves were also drawn using the R packages “timeROC”, “survival”, and “survminer” to evaluate the accuracy of the constructed prediction model.

[0013] S7: Risk score and clinical correlation analysis: The R package “beeswarm” was used to perform risk score and correlation analysis between prognostic metabolic gene expression and clinical characteristics;

[0014] S8: Perform multi-omics analysis of prognosis-related metabolic genes;

[0015] S9: Perform protein interaction network analysis of prognosis-related metabolic genes;

[0016] S10: Perform functional enrichment, pathway and gene set enrichment analysis of prognosis-related metabolic genes;

[0017] S11: Perform drug sensitivity analysis of prognosis-related metabolic genes using CellMiner TM Database and R language, the online database GSCALite integrates the gene expression profiles in GDSC and CTRP and the drug sensitivity data of cancer cell lines for analysis, and then records the data obtained from the relevant analysis, which can complete the entire method of constructing a colorectal cancer metabolism-related prognostic model based on multi-omics.

[0018] In a preferred embodiment, in step S1, colorectal cancer dataset chip data is downloaded from the GEO database, and analyzed in combination with TCGA-CRC data to screen for prognosis-related metabolic genes in colorectal cancer, specifically including:

[0019] First, use the Perl script to get the KEGG dataset:

[0020] Obtain metabolism-related genes in c2.cp.kegg.v7.0.symbols.gmt;

[0021] Secondly, the above metabolic genes were intersected with the gene expression levels in TCGA-CRC to obtain the expression levels of metabolic-related genes in CRC patients; finally, the expression levels of metabolic-related genes commonly expressed in CRC patients in different databases were obtained by intersecting with the gene expression levels in GEO-CRC.

[0022] In a preferred embodiment, in step S2, the selection criteria are: |log2 FC| > 1.5, and False Discovery Rate (FDR) < 0.05. The FC in log2FC stands for foldchange, representing the ratio of expression levels between two samples (groups). Generally, |log2 FC| > 1 is selected as the starting criterion for screening differentially expressed genes; and FDR < 0.05 is generally selected as the default criterion.

[0023] In a preferred embodiment, in step S3, the R software package "ggplot2" is used to generate a volcano plot of differentially expressed genes. Finally, univariate Cox regression analysis is used to screen out metabolic-related genes that are significantly correlated with CRC overall survival (OS) (P < 0.05).

[0024] In a preferred embodiment, in step S4, the R software package "glmnet" is used to fit the LASSO (Least Absolute Shrinkage and Selection Operator) regression to establish an optimal prognostic model for CRC metabolic genes. LASSO regression was first proposed by Robert Tibshirani in 1997. Survival analysis was performed using the Kaplan–Meier method and the log-rank test (P < 0.05). The risk score for each patient was calculated using the following formula:

[0025] Risk Score

[0026] Coefj represents the coefficient, and Xj represents the expression level of each metabolic-related gene in the prognostic model. Survival data for each patient were obtained from CRC clinical information downloaded from the TCGA database (excluding patients with a survival time of less than 90 days). CRC patients were divided into high- and low-risk groups according to their median risk scores. Survival curves for different risk groups were generated using the R packages "survival" and "survminer." Receiver operating characteristic (ROC) curves were generated using the R package "survivalROC" to evaluate the sensitivity and specificity of the metabolic gene risk score survival prediction model. The ROC prediction index was assessed using the area under the curve (AUC). Risk curves, survival status plots, and heat maps were generated using the R language according to the patient's different risk scores. Univariate and multivariate Cox regression analyses were used to identify independent prognostic factors for CRC. Finally, the prognostic model constructed using prognostic metabolic-related genes was validated in two independent GEO colorectal cancer cohorts (GSE39582 and GSE87211).

[0027] In a preferred embodiment, in step S5, the R software packages "Hmisc", "lattice", "Formula", "ggplot2", and "foreign" are used to draw the nomogram and visualize it. Points are assigned according to the percentage contribution of each indicator to the survival risk. The total score for each patient is calculated by summing up the scores of all factors. The survival rate is estimated by drawing a vertical line between each prognostic axis and the total score axis. The nomogram can help doctors make judgments on the prognosis of CRC patients. The consistency between the expected survival rate and the actual survival rate is evaluated using a calibration curve. The consistency index (concordanceindex, C-index) is mainly calculated by the cph function and the validate function in the R package "rms". C-index is often used to evaluate the predictive accuracy of patient prognosis models.

[0028] In a preferred embodiment, in step S7, the R package "ComplexHeatmap" is used to draw a clinical correlation analysis heat map. A chi-square test is performed to observe whether there are differences in clinical traits between the high-risk and low-risk groups, and the R package "RColorBrewer" is used to visualize the clinical correlation analysis results between the high-risk and low-risk groups.

[0029] In a preferred embodiment, in step S8, the multi-omics analysis includes SNV, CNV, Methylation, and Pathway Activity analysis.

[0030] In a preferred embodiment, in step S9, the protein interaction network analysis uses STRING (Interactive Online Gene Search Tool) and cytoscape software to construct a PPI network.

[0031] In a preferred embodiment, in step S10, the R software package "ggplot2", "org.Hs.eg.db", and "clusterProfiler" are used to perform functional enrichment, KEGG analysis, and GESA analysis on prognosis-related metabolism-related genes and their associated genes.

[0032] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are:

[0033] In the present invention, through in-depth research on metabolism-related genes, the metabolic gene-related prognostic model constructed for the first time can more accurately predict the prognosis of colorectal cancer patients. Compared with traditional clinical indicators and pathological characteristics, the present invention utilizes genomic data to make prognostic predictions more reliable and personalized, helping clinicians develop more effective treatment plans. Existing colorectal cancer prognostic models mostly rely on clinical and pathological data and lack comprehensive consideration of metabolism-related genes. The present invention remedies this shortcoming by integrating multi-omics data, providing a new perspective and method, effectively improving the accuracy and practicality of prognostic models, and filling a gap in the existing technology. At the same time, the method of the present invention is not only applicable to the prognosis prediction of colorectal cancer patients, but can also be extended to other types of cancer. Due to the important role of metabolism-related genes in various cancers, the technical solution of the present invention has strong universality and promotion value, and can provide a reference for prognosis prediction and personalized treatment of other cancers. Finally, the prognostic model provided by the present invention can be used as a clinical decision support tool to help doctors make more scientific and accurate decisions during diagnosis and treatment. By identifying high-risk patients early and implementing targeted intervention measures, the patient's prognosis can be significantly improved and the risk of recurrence and metastasis can be reduced. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 Schematic diagram of screening for prognosis-related metabolic genes in colorectal cancer according to the present invention;

[0035] Figure 2 This is the first schematic diagram of the survival analysis performed using the Kaplan–Meier method in the present invention;

[0036] Figure 3 This is a second schematic diagram of survival analysis performed using the Kaplan–Meier method in the present invention;

[0037] Figure 4The third schematic diagram of the survival analysis using the Kaplan–Meier method in the present invention;

[0038] Figure 5 Schematic diagram of nomogram analysis and consistency index analysis in the present invention;

[0039] Figure 6 Schematic diagram of the decision curve analysis structure of the present invention;

[0040] Figure 7 Schematic diagram of functional enrichment analysis of prognosis-related metabolism-related genes in the present invention;

[0041] Figure 8 Schematic diagram of pathway analysis of prognosis-related metabolism-related genes in the present invention;

[0042] Figure 9 Schematic diagram of gene set enrichment analysis of prognosis-related metabolism-related genes in the present invention. DETAILED DESCRIPTION

[0043] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0044] Reference Figure 1-9 ,

[0045] Example:

[0046] A method for constructing a colorectal cancer metabolism-related prognostic model based on multi-omics, the method comprising the following steps:

[0047] Step 1) 1. Obtain RNASeq data of TCGA-CRC from the TCGA database. Download colorectal cancer dataset chip data from the GEO database and analyze it in combination with TCGA-CRC data to screen for prognosis-related metabolic genes in colorectal cancer ( Figure 1 ).

[0048] Step 2) LASSO regression was used to construct a prognosis-related metabolic gene model in colorectal cancer, and Kaplan–Meier method was used for survival analysis ( Figure 2 , Figure 3 , Figure 4 ).

[0049] Step 3) Nomogram analysis and consistency index analysis: The nomogram was used to combine age, sex, TNM stage and the risk score of the constructed model to predict the 1-, 3- and 5-year survival probability of TCGA-CRC patients. Each factor was assigned a proportional score based on the contribution of these indicators to the survival risk. The total score was then added together to obtain the predicted 1-, 3- and 5-year survival probability of the patient. The calibration curve showed that the actual survival time of 1, 3 and 5 years was basically consistent with the predicted survival time. It was also verified using the GEO database, proving that the prediction model had high accuracy ( Figure 5 ).

[0050] Step 4) Decision curve analysis: DCA curves were drawn using the R package “ggDCA”. Clinically relevant ROC curves were also drawn using the R packages “timeROC”, “survival”, and “survminer” to evaluate the accuracy of the constructed prediction model ( Figure 6 ).

[0051] Step 5) Correlation Analysis between Risk Scores and Clinical Characteristics: The R package "beeswarm" was used to analyze the correlation between risk scores, prognostic metabolic gene expression, and clinical characteristics. The R package "ComplexHeatmap" was used to create heat maps for the clinical correlation analysis. Chi-square tests were used to examine whether clinical traits differed between high- and low-risk groups. The R package "RColorBrewer" was used to visualize the clinical correlation analysis results between high- and low-risk groups.

[0052] Step 6) Multi-omics analysis of prognosis-related metabolic genes: SNV percentage: The SNV percentage of each gene coding region was calculated by the following formula: number of mutation samples / number of cancer samples. The SNV summary and tumor waterfall plot were generated by the R package "maftools". CNV statistics were based on GISTIC-processed data, and the correlation between CNV and mRNA expression was based on the CNV raw data. The Persons product-moment correlation coefficient was used, and the t-distribution was followed to explore the association between paired mRNA expression and methylation. Gene expression was divided into two groups (high expression group and low expression group) according to the median. The difference in pathway activity score (PAS) between the groups was determined by the student T test, and the P value was adjusted by FDR. FDR < 0.05 indicated a significant difference. When PAS (gene A high expression group) > PAS (gene A low expression group), it is considered that gene A may have an activating effect on the pathway, otherwise it has an inhibitory effect on the pathway.

[0053] Step 7) Protein interaction network analysis of prognosis-related metabolic genes: STRING (interactive online gene search tool) and cytoscape software were used to construct the PPI network.

[0054] Step 8) Functional enrichment analysis of prognosis-related metabolic genes: The functional enrichment method mainly uses Gene Ontology (GO), which explores the research in the field of biology from the following three directions: 1. Biological Process (BP); 2. Cellular Component (CC); 3. Molecular Function (MF) ( Figure 7 ).

[0055] Step 9) Pathway analysis of prognosis-related metabolic genes: The signal pathway analysis in this study was mainly based on the Kyoto Encyclopedia of Genes and Genomes (KEGG) ( Figure 8 ).

[0056] Step 10) Gene Set Enrichment Analysis of Prognostic-Related Metabolic Genes: GSEA was used to reveal the signaling pathways and biological processes of differentially expressed genes between high-risk and low-risk subgroups. GSEA analysis was performed using GSEA_4.1.0 software and Perl software. The number of permutations was set to 1,000, and an FDR < 0.25 indicated statistical significance. Figure 9 ).

[0057] Step 11) Prognosis-related metabolic gene drug sensitivity analysis: mainly based on CellMiner TM The database uses R language and the online database GSCALite to integrate the gene expression profiles in GDSC and CTRP and the drug sensitivity data of cancer cell lines for analysis.

[0058] From the above we can know:

[0059] Through in-depth research on metabolism-related genes, the present invention has constructed a metabolic gene-related prognostic model for the first time, which can more accurately predict the prognosis of colorectal cancer patients. Compared with traditional clinical indicators and pathological features, the present invention utilizes genomic data, making prognostic predictions more reliable and personalized, helping clinicians develop more effective treatment plans. Existing colorectal cancer prognostic models mostly rely on clinical and pathological data and lack comprehensive consideration of metabolism-related genes. The present invention addresses this shortcoming by integrating multi-omics data, providing a new perspective and method, effectively improving the accuracy and practicality of prognostic models and filling a gap in the existing technology. Furthermore, the method of the present invention is not only applicable to the prognosis prediction of colorectal cancer patients, but can also be extended to other types of cancer. Due to the important role of metabolism-related genes in various cancers, the technical solution of the present invention has strong universality and dissemination value, and can provide a reference for prognosis prediction and personalized treatment of other cancers. Finally, the prognostic model provided by the present invention can serve as a clinical decision support tool, helping doctors make more scientific and accurate decisions during diagnosis and treatment. By identifying high-risk patients early and implementing targeted interventions, patients' prognosis can be significantly improved and the risk of recurrence and metastasis can be reduced.

[0060] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprises" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device that includes a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article or device. In the absence of further limitations, an element defined by the phrase "comprises a ..." does not exclude the presence of other identical elements in the process, method, article or device that includes the element.

[0061] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A method for constructing a multi-omics-based metabolism-related prognostic model for colorectal cancer, characterized by: The method comprises the following steps: S1: RNASeq data of TCGA-CRC were obtained from the TCGA database; S2: Differential expression analysis was performed: Wilcox test and R package "edgeR" were used to screen differentially expressed metabolic genes in CRC; S3: Screening of differentially expressed genes related to prognosis; S4: Construction of prognosis-related metabolic gene model and survival analysis in colorectal cancer; S5: Perform nomogram analysis and consistency index: Use the nomogram combined with clinicopathological characteristics (TNM stage, pathological stage, age, sex) and risk score to predict the 1-, 3-, and 5-year overall survival of CRC patients; S6: Decision curve analysis was performed: DCA curves were drawn using the R package "ggDCA" and clinically relevant ROC curves were drawn using the R packages "timeROC", "survival", and "survminer" to evaluate the accuracy of the constructed prediction model; S7: Risk score and clinical correlation analysis: The R package "beeswarm" was used to perform risk score and correlation analysis between prognostic metabolic gene expression and clinical characteristics; S8: Perform multi-omics analysis of prognosis-related metabolic genes; S9: Perform protein interaction network analysis of prognosis-related metabolic genes; S10: Perform functional enrichment, pathway and gene set enrichment analysis of prognosis-related metabolic genes; S11: Perform drug sensitivity analysis of prognosis-related metabolic genes using CellMiner TM Database and R language, the online database GSCALite integrates the gene expression profiles in GDSC and CTRP and the drug sensitivity data of cancer cell lines for analysis, and then records the data obtained from the relevant analysis, which can complete the entire method of constructing a colorectal cancer metabolism-related prognostic model based on multi-omics.

2. The method for constructing a multi-omics-based colorectal cancer metabolism-related prognostic model according to claim 1, characterized in that: In step S1, colorectal cancer dataset chip data is downloaded from the GEO database, and analyzed in combination with TCGA-CRC data to screen for prognosis-related metabolic genes in colorectal cancer, specifically including: First, a Perl script was used to extract the KEGG dataset; Obtain metabolism-related genes in c2.cp.kegg.v7.0.symbols.gmt; Secondly, the above metabolic genes were intersected with the gene expression levels in TCGA-CRC to obtain the expression levels of metabolic-related genes in CRC patients; finally, the expression levels of metabolic-related genes commonly expressed in CRC patients in different databases were obtained by intersecting with the gene expression levels in GEO-CRC.

3. The method for constructing a multi-omics-based colorectal cancer metabolism-related prognostic model according to claim 1, characterized in that: In step S2, the selection criteria are: |log2FC|>1.5, false discovery rate, FC in log2FC stands for foldchange, which represents the ratio of expression levels between two samples; generally, |log2 FC|>1 is selected as the starting standard for screening differentially expressed genes; and generally, FDR<0.05 is selected as the default standard.

4. The method for constructing a multi-omics-based colorectal cancer metabolism-related prognostic model according to claim 1, characterized in that: In step S3, the R software package "ggplot2" was used to generate a volcano plot of differentially expressed genes; finally, univariate Cox regression analysis was used to screen out metabolism-related genes that were significantly correlated with CRC overall survival.

5. The method for constructing a multi-omics-based colorectal cancer metabolism-related prognostic model according to claim 1, characterized in that: In step S4, the R software package "glmnet" was used to fit the LASSO regression to establish the optimal prognostic model of CRC metabolic genes; LASSO regression was first proposed by Robert Tibshirani in 1997; survival analysis was performed using the Kaplan–Meier method and the log-rank test; The risk score for each patient was calculated using the following formula: Coef j represents the coefficient, X j represents the expression level of each metabolic-related gene in the prognostic model; the survival data of each patient were obtained from the CRC clinical information downloaded from the TCGA database; CRC patients were divided into high-risk and low-risk groups according to the median risk value; the R packages "survival" and "survminer" generated survival curves for different risk groups; the receiver operating characteristic curve was drawn by the R package "survival ROC", and the ROC curve was used to evaluate the sensitivity and specificity of the metabolic gene risk score survival prediction model; the ROC prediction index was evaluated by the area under the curve; risk curves, survival status diagrams and heat maps were generated using R language according to the different risk scores of patients; independent prognostic factors for CRC were determined by univariate and multivariate Cox regression analysis methods; finally, the prognostic model constructed by prognostic metabolic-related genes was validated in two independent GEO colorectal cancer cohorts (GSE39582 and GSE87211).

6. The method for constructing a multi-omics-based colorectal cancer metabolism-related prognostic model according to claim 1, characterized in that: In step S5, the R software packages "Hmisc", "lattice", "Formula", "ggplot2", and "foreign" are used to draw the nomogram and visualize it; each indicator is scored according to its contribution percentage to the survival risk; A total score was calculated for each patient by summing the scores of all factors; Survival rates were estimated by drawing a vertical line between each prognostic axis and the total score axis. A nomogram was used to help physicians make judgments about the prognosis of CRC patients. A calibration curve was used to assess the consistency between expected and actual survival rates. The consistency index was calculated using the cph and validate functions in the R package "rms." The C-index was often used to assess the predictive accuracy of patient prognostic models.

7. The method for constructing a multi-omics-based colorectal cancer metabolism-related prognostic model according to claim 1, characterized in that: In step S7, the R package "ComplexHeatmap" draws a clinical correlation analysis heat map; a chi-square test is used to observe whether there are differences in clinical traits between the high-risk and low-risk groups, and the R package "RColorBrewer" is used to visualize the clinical correlation analysis results between the high-risk and low-risk groups.

8. The method for constructing a multi-omics-based colorectal cancer metabolism-related prognostic model according to claim 1, characterized in that: In step S8, the multi-omics analysis includes SNV, CNV, Methylation, and Pathway Activity analysis.

9. The method for constructing a multi-omics-based colorectal cancer metabolism-related prognostic model according to claim 1, characterized in that: In step S9, the protein interaction network analysis uses STRING and cytoscape software to construct a PPI network.

10. The method for constructing a colorectal cancer metabolism-related prognostic model based on multi-omics according to claim 1, characterized in that: In step S10, the R software package "ggplot2", "org.Hs.eg.db", "clusterProfiler" is used to perform functional enrichment, KEGG analysis, and GESA analysis on prognosis-related metabolism-related genes and their associated genes.

Citation Information

Cited By

  • Colorectal cancer liver metastasis detection method and system based on multi-omics deep learning

    CN121281845A