Hepatocellular carcinoma gene knockout target library based on multiple omics and screening method thereof

Abstract

The invention relates to a hepatocellular carcinoma gene knockout target library based on multiple omics and a screening method of the hepatocellular carcinoma gene knockout target library, and the gene knockout target library for precise treatment of hepatocellular carcinoma is finally obtained through data collection and integration, data screening and target verification in sequence. According to the invention, through multi-omics data integration and bioinformatics analysis, key driving genes of hepatocellular carcinoma are systematically screened, and the important effects of the genes in occurrence, development, metastasis, drug resistance and immune escape of hepatocellular carcinoma are disclosed; the genes not only deepen the understanding of the hepatocellular carcinoma molecular mechanism, but also provide important theoretical basis and potential intervention targets for the development of targeted therapy and personalized therapy strategies.

Description

Technical Field

[0001] This invention relates to a multi-omics-based hepatocellular carcinoma gene knockout target library and its screening method, belonging to the field of medical bioinformatics technology. Background Technology

[0002] Hepatocellular carcinoma (HCC) is one of the most common malignant tumors worldwide, with persistently high incidence and mortality rates. Its incidence is significantly higher in Asia and Africa than in other regions. Statistics show that liver cancer is the third leading cause of cancer-related deaths globally, with HCC accounting for the vast majority of cases (approximately 75%-85%). The high mortality rate of HCC is closely related to the difficulty of early diagnosis. Due to the liver's strong compensatory function, early-stage HCC patients often have no obvious symptoms, leading to many patients being diagnosed at an advanced stage, missing the optimal treatment window. More than 70% of patients are diagnosed at an advanced stage, often accompanied by liver failure, portal hypertension, or distant metastasis, further limiting treatment options and reducing survival rates.

[0003] Despite recent advancements in the diagnosis and treatment of liver cancer (HCC), such as improvements in imaging techniques, the application of serum biomarkers (e.g., alpha-fetoprotein (AFP), and the widespread use of targeted therapies (e.g., sorafenib and lenvatinib), existing treatment methods remain limited due to the high heterogeneity and complex molecular mechanisms of HCC. Surgical resection and liver transplantation, while the preferred treatments for early-stage HCC, are only suitable for a subset of patients, with postoperative recurrence rates as high as 50%-70%. Chemotherapy and radiotherapy have limited efficacy in advanced-stage patients and significant side effects. While targeted therapy and immunotherapy offer new hope for advanced-stage patients, their efficacy is still limited by tumor heterogeneity and drug resistance. Statistics show that approximately 60% of locally resectable patients survive 5 years after diagnosis, while the 5-year survival rate for patients with distant metastases is only 4%, further highlighting the challenges of HCC treatment. Furthermore, the high recurrence and metastasis rates of HCC are closely related to its microenvironment, immune escape mechanisms, and epigenetic alterations; these complex biological characteristics make HCC treatment even more difficult. Therefore, in-depth research into the molecular mechanisms of liver cancer, the development of novel therapeutic targets, and the optimization of existing treatment strategies are of significant clinical importance.

[0004] In recent years, with the rapid development of multi-omics technologies (such as genomics, transcriptomics, proteomics, and metabolomics) and bioinformatics algorithms, systematic screening of key genes closely related to the occurrence, development, and metastasis of liver cancer has become possible. For example, target screening based on multi-omics data can be combined with drug databases (such as DrugBank and ChEMBL) to predict new indications for existing drugs, thereby accelerating the development of liver cancer drugs. Furthermore, functional studies based on gene editing technologies such as CRISPR-Cas9 have provided powerful tools for elucidating the molecular mechanisms of liver cancer and validating targets.

[0005] Gene knockout technology is an experimental method that uses artificial intervention to disable specific genes, providing a powerful tool for studying gene function, elucidating disease mechanisms, and developing new treatment strategies. Since homologous recombination technology was first applied to gene knockout in the 1980s, this field has experienced rapid development. In recent years, in particular, the emergence of CRISPR-Cas9 technology has revolutionized the field of gene editing. Its system originates from the adaptive immune mechanism of bacteria, and by designing specific guide RNAs (gRNAs), it can precisely target the target gene sequence and introduce a double-strand break (DSB) at a specific location. This allows for gene knockout, insertion, or modification through non-homologous end-joining (NHEJ) or homology-directed repair (HDR) mechanisms. Compared to traditional gene knockout technologies (such as zinc finger nucleases (ZFNs) and transcription activator-like effector nucleases (TALENs), CRISPR-Cas9 offers advantages such as ease of operation, low cost, high efficiency, and the ability to simultaneously target multiple genes, thus rapidly becoming a mainstream tool in functional genomics research.

[0006] In liver cancer research, CRISPR-Cas9 provides a powerful tool for elucidating the molecular mechanisms of liver cancer and developing new therapeutic strategies. Its high efficiency, precision, and programmability have made it a significant breakthrough in functional genomics research. For example, using CRISPR-Cas9 technology, researchers can systematically knock out candidate genes in liver cancer cells, observing their effects on cell proliferation, apoptosis, migration, and invasion phenotypes, thereby screening for key genes closely related to liver cancer. Furthermore, gene knockout technology can also be used to construct animal models of liver cancer. For instance, by knocking out specific genes (such as p53 or PTEN) in the mouse liver, the process of human liver cancer development can be simulated, providing an important experimental platform for studying the molecular mechanisms of liver cancer and drug screening.

[0007] However, despite the immense potential of CRISPR-Cas9 technology in gene function research, efficiently and accurately screening key gene targets associated with liver cancer remains a major challenge. First, liver cancer is a highly heterogeneous tumor with complex molecular mechanisms involving the regulation of multiple signaling pathways and gene networks. Traditional single-gene knockout experiments struggle to fully reflect the complexity of liver cancer and are time-consuming and costly. Second, existing target screening methods largely rely on single-omics data (such as the genome or transcriptome), lacking integrated analysis of multi-dimensional data, which can lead to biases or omissions in the screening results. For example, some genes may not show significant mutations at the genomic level but exhibit significant differences at the transcriptome or proteome levels; these genes may be overlooked by traditional methods. Furthermore, the redundancy and compensatory mechanisms of gene function also increase the difficulty of target screening; the knockout of some genes may be compensated for by the function of other genes, thus masking their true biological role.

[0008] To overcome these challenges, researchers need to develop more efficient and precise target screening strategies. For example, combining multi-omics data and bioinformatics algorithms can systematically screen key genes associated with liver cancer, revealing the molecular mechanisms of liver cancer more comprehensively and identifying important targets that may be overlooked in single-omics data. Furthermore, using high-throughput CRISPR screening technology, thousands of genes can be knocked out simultaneously, and combined with functional phenotypic analysis, genes closely related to the liver cancer phenotype can be quickly screened; this high-throughput screening method not only improves experimental efficiency but also discovers new potential therapeutic targets.

[0009] There is an urgent need to develop a gene knockout target library for hepatocellular carcinoma based on multi-omics data and its algorithm process, which is also one of the important directions in the current field of liver cancer research. Summary of the Invention

[0010] The main objective of this invention is to overcome the problems existing in the prior art and propose a multi-omics-based method for screening hepatocellular carcinoma (HCC) gene knockout target libraries. This method can systematically screen key driver genes of HCC and reveal the important roles of these genes in the occurrence, development, metastasis, drug resistance, and immune escape of HCC. Furthermore, the invention proposes a HCC gene knockout target library obtained by the above method. These genes not only deepen the understanding of the molecular mechanisms of HCC but also provide important theoretical basis and potential intervention targets for the development of targeted therapy and personalized treatment strategies.

[0011] The technical solution of this invention to solve its technical problem is as follows:

[0012] A multi-omics-based method for screening gene knockout target libraries for hepatocellular carcinoma includes the following steps:

[0013] Step 1: Data Collection and Integration

[0014] S1. Search public databases to collect reported gene information data related to hepatocellular carcinoma, and after systematic analysis, draw a map of the core driver genes of hepatocellular carcinoma and the key signaling pathways they regulate.

[0015] S2. Systematically organize and analyze gene mutation data related to hepatocellular carcinoma in published literature, including: obtaining gene mutation data in protein-coding regions by organizing and analyzing whole-exome sequencing results; obtaining genome-wide variation data, including variation data in exon regions, intron regions, and regulatory regions, by organizing and analyzing whole-genome sequencing results.

[0016] S3. Based on RNA sequencing data, the DESeq2 software tool in R language was used to analyze and identify differentially expressed genes (DEGs) between tumor and normal tissues.

[0017] The second step is data screening: After summarizing all the genes obtained in the first step, risk ratio analysis, survival analysis, different tumor stage analysis, UALCAN-based analysis, and research activity assessment analysis based on the number of literatures are performed. Then, all the analysis results are summarized to obtain a candidate gene knockout target library for precision treatment of hepatocellular carcinoma.

[0018] Step 3, Target Validation: For the candidate gene knockout target library for precision treatment of hepatocellular carcinoma obtained in Step 2, the expression level and expression pattern of each gene in different hepatocellular carcinoma cell lines are analyzed, and genes that are not expressed in all hepatocellular carcinoma cell lines are removed; finally, a gene knockout target library for precision treatment of hepatocellular carcinoma is obtained.

[0019] This method first identifies core driver genes and key regulatory pathways in hepatocellular carcinoma (HCC) through literature mining; then, it uses multi-omics sequencing data to map HCC mutations and screens differentially expressed genes based on multiple independent transcriptome datasets; hazard ratio and survival analysis methods are introduced to comprehensively evaluate the clinical relevance of candidate genes; simultaneously, a correlation model between gene expression profiles and tumor stage, grade, and metastatic status is constructed to identify key targets with diagnostic and therapeutic potential. Furthermore, a validation scheme for HCC cancer cell line expression profiles is provided, achieving a complete closed loop from algorithm prediction to experimental validation, providing intelligent decision support for precision treatment of HCC.

[0020] The further improved technical solution of this invention is as follows:

[0021] Preferably, in step S1, the public database includes the PubMed database of the National Center for Biotechnology Information (NCBI); the gene information data related to hepatocellular carcinoma includes: genes, signaling pathways, mutation profiles, expression profiles, epigenetic modifications, protein-protein interaction networks, and metabolic pathway data; the atlas includes: cell cycle regulation, telomere maintenance, WNT / β-catenin signaling pathway, chromatin remodeling, hepatocyte differentiation, RAS / MAPK / PI3K / AKT / mTOR signaling pathway, TGF-β signaling pathway, anti-oxidative stress resistance, epigenetic regulation, receptor tyrosine kinase (RTK) signaling pathway, RAS / RAF / MEK / ERK signaling pathway, JAK-STAT signaling pathway, Hedgehog signaling pathway, Hippo signaling pathway, apoptosis pathway, NOTCH signaling pathway, and hypoxia signaling pathway. The signaling pathways include the Brachyury signaling pathway and the FGF signaling pathway. Note: Abnormal activation or dysfunction of these pathways plays a key role in the occurrence, progression, metastasis, drug resistance, and immune escape of hepatocellular carcinoma.

[0022] By adopting the above preferred scheme, the specific technical details of the first step S1 can be further optimized.

[0023] Preferably, in step S2, the variation data specifically includes: single nucleotide variants (SNPs), insertions / deletions (Indels), copy number variations (CNVs), and structural variations (SVs).

[0024] Preferably, the specific process of step S2 in the first step includes:

[0025] S2-1. Existing literature conducted a multi-omics integrated analysis of 257 primary tumor regions and 176 metastatic tumor regions from 182 hepatocellular carcinoma patients. Based on this existing literature, a systematic review and analysis were conducted, and the following hepatocellular carcinoma mutated genes were found to be commonly found in the Cancer Genome Atlas (TCGA): TP53, CTNNB1, ALB, AXIN1, BAP1, KEAP1, NFE2L2, LZTR1, RB1, PIK3CA, RPS6KA3, AZIN1, KRAS, IL6ST, RP1L1, CDKN2A, EEF1A1, ARID2, ARID1A, GPATCH4, ACVR2A, APOB, CREB3L3, NRAS, AHCTF1, HIST1H1C, PTEN, and BRD7.

[0026] S2-2. Existing literature was integrated to collect genomic data from 1340 hepatocellular carcinoma cases from different ethnic groups. Based on this existing literature, a systematic review and analysis were conducted, and it was concluded that the TP53, TERT, and WNT (CTNNB1) signaling pathways are the three core driving factors. Low-frequency mutated genes involved in multiple cancer signaling pathways were identified as follows: ALB, AXIN1, ARID1A, ARID2, NFE2L2, CDKN2A, RPS6KA3, RB1, ACVR2A, FGF19, CCND1, APOB, EEF1A1, BRD7, PTEN, PIK3CA, TSC2, BAP1, CDKN2B, KEAP1, KRTAP5-11, CDKN1A, KRAS, RPL22, NBEA, PCF11, TSC1, HNF1A, FGA, ERRF11.

[0027] S2-3. Existing literature has systematically studied the frequency of genomic alterations in 26 patients with hepatocellular carcinoma through circulating tumor DNA analysis. Based on the systematic review and analysis of this existing literature, the mutated genes identified include: TP53, CTNNB1, ARID1A, EGFR, MYC, APC, ATM, CDK6, ERBB2, MET, BRCA1, CCNE1, CDKN2A, FGFR1, FGFR2, KIT, KRAS, NF1, NFE2L2, PIK3CA, RAF1, ALK, AR, BRAF, BRCA2, CCND1, CCND2, CDK4, ESR1, JAK3, MAP2K1, NTRK1, PTEN, RB1, RET, SMAD4, and TSC1.

[0028] S2-4. Analysis of existing literature: The GSE273254 dataset in the public database GEO was analyzed. Based on this existing literature, a systematic organization and analysis were conducted, and the mutated genes identified included: TP53, TERT, LRP1B, CTNNB1, AXIN1, ARID1A, ATM, TSC2, NOTCH3, APC, KMT2A, ARID2, ROS1, NTRK3, ERBB4, KMT2D, NF1, KMT2C, BCOR, KDR, NOTCH4, ATR, STAT3, NCOR1, EP300, IRF2, HSPD1, KEAP1, TSHR, SYK, PAX5, PDGFRA, and BAP1.

[0029] S2-5. Summarize the mutated genes obtained from the above steps to obtain gene mutation data related to hepatocellular carcinoma.

[0030] By adopting the above preferred scheme, the specific technical details of the first step S2 can be further optimized.

[0031] More preferably, the specific process of step S3 includes:

[0032] S3-1. Existing research has constructed survival prediction models for immune-related genes based on the GSE214846 database. One study selected 65 liver cancer patients and their paired adjacent normal tissues for RNA sequencing and gene expression profiling analysis. Based on the RNA sequencing data from this existing study, differentially expressed genes were first screened using the DESeq2 software, including: DKK1, TERT, SFN, NQO1, CDK1, E2F1, NKD1, CDKN2A, SLC2A1, EF1, HDAC11, TXNRD1, IGF2, E2F3, ACACA, BAK1, SQSTM1, CDK4, SKP2, and PLCG1.

[0033] S3-2. Utilizing existing cancer genome maps and the GTEx genotype tissue expression database, gene expression data from 421 hepatocellular carcinoma cases and 110 normal tissue cases were extracted from this database. Differentially expressed genes were first screened using the DESeq2 software, including: REG3A, GPC3, NQO1, AFP, MDK, UBD, AKR1B10, SPP1, IFI27, GLUL, LCN2, IGF2, HUL C, CD74, H19, HLA-A, PABPC1, LGALS3BP, GRN, SQSTM1, RP11-386G11.10, ISG15, S100A6, GGH, SPINK1, HLA-F, HLA-B, CTSA, CLIC1, and TM4SF5.

[0034] S3-3. Using the existing GSE248562 dataset, RNA sequencing data from eight HepG2 samples and eight L02 samples were selected from this dataset. The differentially expressed genes were first screened using the DESeq2 software, including: APOA1, NR0B2, TTR (transferrin), SERPINA1, NDRG2, ALDH1A1, CES1, APOC3, FABP1, TF (transferrin), FGA, SERPINA1, APOE, FGG, ALB, HNF4A, ANG (angiogenin), DUSP6, CXCL16 and FN1.

[0035] S3-4. Utilize the existing GSE157905 project; select data from four samples (SNU449-CT, SNU449-Gefitinib, SNU449-Lenvatinib, SNU449-Combination, Huh6), four samples (SNU182), and four samples (JHH1) from this project. First, use the software DESeq2 to analyze and screen differentially expressed genes, including: CDH2, SPARC, VIM, MMP2, POSTN, APOE, FN1, FGF2, COL1A1, COL1A2, LOXL2, AKT3, LOX, NRP1, TGFB1, SERPINE1, WN T5A, CAV1, CMBL, and IRS1.

[0036] S3-5. Using the existing GSE117623 dataset, data from 52 hepatocellular carcinoma cases and 65 normal liver tissue samples were selected from this project. The differentially expressed genes were first screened using the DESeq2 software, including: CA1, OSBP2, TRIM58, MXI1, PRDX2, MAP2K3, SELENBP1, ADIPOR1, BLVRB, HBD, RBM38, BAG1, BNIP3L, FOXO3, BSG, CTSB, BCL2L1, DUSP1, EIF5, and HK1.

[0037] S3-6. Using the existing Cancer Cell Line Encyclopedia (CCLE) database, we screened out the VIM high-expression groups JHH2, SNU449, and SKHEP1, and the VIM low-expression groups JHH6, HEPG2, and JHH5. Based on their data, we first used the DES eq2 software to analyze and screen out differentially expressed genes, including: LOXL, CXCL1, EFEMP1, VIM, MYL9, BGN, TNC, IL1B, PMP22, COL1A1, CCL2, SRI, ETS1, FSTL1, SRPX, ITGA5, CXCL8, VCAN, VLDLR, and PMEPA1.

[0038] S3-7. Using the existing GSE186191 dataset, RNA sequencing data of lenvatinib-resistant liver cancer cell models Hep3B-LR and Huh7-LR, and corresponding parental cells Hep3B-P and Huh7-P, were selected from this dataset. Differentially expressed genes were first screened using the DESeq2 software, including: CCNG2, RRAGD, ME1, LDHA, PPIA, ME2, SERPINE1, PPP1R15A, DSC2, ZNF292, KDELR3, NAGK, SLC16A3, GMPPA, LGALS3, HMMR, JMJD6, MMP14, IER3, MMP15, TP53, and HDAC2.

[0039] By adopting the above preferred scheme, the specific technical details of the first step S3 can be further optimized.

[0040] Preferably, in the second step, the specific process of the risk ratio analysis includes:

[0041] Clinical data of hepatocellular carcinoma (HCC) patients and gene expression data for each HCC patient were obtained from the Cancer Genome Atlas (TCGA). Cox regression analysis was performed using the R software tools survival and survminer to calculate the hazard ratio (HR) of each gene. Genes with the highest HR values ​​were selected from the analysis results, including: TBXT, WNT8A, CSNK1A1L, TERC, PIK3R2, GSTT4, ATR, WNT16, NTRK1, ALK, EPB42, RP1L1, BDNF, ATM, TSHR, APC2, SEM1, PRKCG, SNCA, TRIM58, SF3B1, KMT2D, MYB, PPIA, SMARCD1, COL1A2, AKT2, SMAD2, POLK, FUBP1, ZNF292, HSPD1, AXIN1, CNGA3, PCF11, HULC, GSTA3, MK RN1, FUS, and CDK2.

[0042] Preferably, in the second step, the specific process of the survival analysis includes:

[0043] Kaplan-Meier survival curves were plotted to assess the correlation between gene expression levels and overall survival (OS) and relapse-free survival (RFS). During the analysis, patients were automatically categorized into high-expression and low-expression groups by selecting the optimal cutoff value. After analysis, OS and RFS results for each gene were recorded, with particular attention paid to genes with high expression and short OS or RFS. Statistical significance was assessed by recording the p-value for each gene to determine whether the correlation between its expression level and OS and RFS was statistically significant. Genes identified as statistically significant included ACACA and ATR. , AXIN1, BRCA1, BSG, CCNE1, CDK1, CDK2, CDK4, CDK5, CDKN2A, CDKN2D, CLIC1, CTAG1B, E2F1, E2F2, E2F3, EIF4E, EWSR1, EZH2, FUBP1, G PATCH4, GRB2, GRN, HIST1H3C, HMMR, HRAS, HSPD1, ME2, MMP1, MYB, OSBP2, PAX8, PLCB1, PPIA, POLD1, SFN, SMARCD1, SNRPE, SPP1, SQST M1, SRI, SSRP1, STC1, TERT, TNNT1, TSC1, TXNRD1, VEGFA.

[0044] Preferably, in the second step, the specific process of analyzing different tumor stages includes:

[0045] Clinical data of hepatocellular carcinoma patients were obtained from the Cancer Genome Atlas (TCGA), along with RNA sequencing expression data containing the expression levels of each gene in each patient. Then, the differences in the expression levels of each gene in different tumor stages I-IV, as well as in early stages I-II and late stages III-IV, were analyzed.

[0046] The specific process of the UALCAN-based analysis includes:

[0047] A systematic analysis of gene expression in hepatocellular carcinoma was conducted, including: Sample type analysis: comparing the expression differences of each gene in normal liver tissue and hepatocellular carcinoma tissue to preliminarily determine their activation or inhibition during carcinogenesis; Cancer staging analysis: analyzing the expression differences of each gene in different tumor stages I-IV and assessing their correlation with tumor progression; Tumor grading analysis: analyzing the expression differences of each gene in different tumor grades 1-4 and assessing their correlation with tumor differentiation and malignancy; Lymph node metastasis status analysis: analyzing the expression differences of each gene in patients with and without lymph node metastasis to study their regulatory role in tumor metastasis.

[0048] The specific process of the research activity assessment and analysis based on the number of documents includes:

[0049] Using the PubMed database of the National Center for Biotechnology Information (NCBI), we retrieved literature on each gene in the field of hepatocellular carcinoma research and counted the number of related literatures to determine the research activity of the gene in the field of liver cancer.

[0050] Based on the combined results of the three analyses above, the identified genes include: AKT1, ATM, ATR, AXIN1, AXIN2, BAK1, BRCA1, BSG, CCNE1, CDK1, CDK2, CDK5, CDK6, CDKN2B, DNMT3A, E2F3, EIF4E, EPCAM, EZH2, FGFR1, FOXO3, FZD7, GLI1, HDAC2, HMMR, HMOX1, HRAS, IDH1, IGF1R, IL1B, KIT, KRAS, MUC1, MYB, NOTCH1, PDK1, POSTN, PPARG, RB1, SALL4, SERPINE1, SMAD2, SRI, TGFBR1, TSC1, and YAP1.

[0051] By adopting the above preferred solutions, the specific technical details of the second step can be further optimized.

[0052] Preferably, in the third step, relevant data from the CCLE (Encyclopedia of Cancer Cell Lines) database is used for analysis.

[0053] This invention also provides:

[0054] A gene knockout target library for precision treatment of hepatocellular carcinoma obtained by the screening method described above.

[0055] This invention systematically screens key driver genes for hepatocellular carcinoma (HCC) through multi-omics data integration and bioinformatics analysis, revealing their crucial roles in HCC occurrence, development, metastasis, drug resistance, and immune escape. These genes not only deepen our understanding of the molecular mechanisms of HCC but also provide important theoretical basis and potential intervention targets for the development of targeted therapy and personalized treatment strategies. This invention integrates multi-dimensional omics data, including genomic mutation maps, transcriptomic differential expression profiles, and protein-protein interaction networks, combined with bioinformatics algorithms and clinical prognostic data, to systematically screen key genes closely related to the occurrence, development, and metastasis of HCC. This significantly improves the accuracy and reliability of target screening, providing an important technical platform and decision support tool for precision treatment and drug development in HCC. The application of this invention will contribute to advancing precision medicine research in HCC, providing patients with more effective treatment options, and has broad clinical application prospects and socio-economic benefits. Attached Figure Description

[0056] Figure 1 The map shows the core driver genes of hepatocellular carcinoma and the multiple key signaling pathways they regulate, as illustrated in Example 1 of this invention.

[0057] Figure 2 This is a graph showing the gene mutation frequency results of 40 samples in the GSE273254 dataset in Example 1 of the present invention.

[0058] Figure 3 This is a schematic diagram illustrating the data related to five candidate targets for gene knockout in hepatocellular carcinoma, as exemplified in Embodiment 2 of the present invention: the correlation between gene expression levels and overall survival (OS) and recurrence-free survival (RFS); the expression levels of patients in TNM (Tumor, Node, Metastasis) stages (stages I-IV), early (stages I-II), and late (stages III-IV); the expression levels in different differentiation grades; the expression levels in normal liver and primary tumor tissue; and the correlation with lymph node metastasis status (N stage: normal, N0 without metastasis, N1 with metastasis); the prognostic risk assessment of each gene: the hazard ratio; and the number of research papers related to hepatocellular carcinoma in the PubMed database for each target.

[0059] Figure 4 This is a schematic diagram of the overall process of the present invention. Detailed Implementation

[0060] The present invention will be further described in detail below with reference to the embodiments. However, the present invention is not limited to the examples given.

[0061] Example 1

[0062] This example illustrates data collection and integration.

[0063] The specific details of this embodiment are as follows:

[0064] 1.1 Literature Search

[0065] By searching the PubMed database of the National Center for Biotechnology Information (NCBI), we systematically collected reported gene and signaling pathway information related to hepatocellular carcinoma. The search results are shown in Table 1, and further plotting was performed based on these results. Figure 1 This study aims to showcase the core driver genes of hepatocellular carcinoma and the key signaling pathways they regulate, providing important theoretical basis and visual reference for subsequent research, as detailed below.

[0066] 1) Cell Cycle Regulation: Key genes include TP53, ATM, RB1, CDKN2A, MYC, and CCNE1. These genes play a central regulatory role in cell division and proliferation. Among them, mutations in TP53 can lead to cell cycle dysregulation and promote the development of hepatocellular carcinoma.

[0067] 2) Telomere Maintenance: TERT is the main regulatory gene for telomere elongation. Its mutation or abnormal activation can enable liver cancer cells to gain unlimited proliferative capacity.

[0068] 3) WNT / β-catenin signaling pathway: Key genes include CTNNB1, AXIN1 / 2, APC, NCOR1, WT1, DVL, CK1α, and GSK3β. Among them, mutations in CTNNB1 can activate the WNT pathway, promote cell growth and proliferation, and drive the development of liver cancer.

[0069] 4) Chromatin Remodeling: Key genes include ARID1A, ARID2, KMT2A / C / B, BAP1, and ARID1B. Mutations in these genes can affect gene expression regulation, leading to the progression and worsening of hepatocellular carcinoma.

[0070] 5) Hepatocyte Differentiation: The key genes ALB and APOB are crucial for normal hepatocyte differentiation. Mutations in these genes may impair liver function and promote the development of hepatocellular carcinoma.

[0071] 6) RAS / MAPK / PI3K / AKT / mTOR signaling pathway: Key genes include RPS6KA3, FGF3 / 4 / 19, EEF1A, PTEN, PIK3CA, TSC1 / 2, KRAS, PDGFRA, EGFR, RPS6KA3, FAK, c-MET, RHEB, S6, 4EBP, eIF4E, and RAP. This pathway regulates the growth, proliferation, and survival of tumor cells, and its mutations are relatively common in hepatocellular carcinoma.

[0072] 7) TGF-β signaling pathway: Key genes ACVR2A and ERK play a role in regulating cell growth and differentiation, and their abnormal activation may promote cancer progression and metastasis.

[0073] 8) Anti-oxidative stress resistance: key genes NFE2L2, KEAP1 and NRF2. Mutations in the NRF2 signaling pathway can enable liver cancer cells to adapt to oxidative stress environment, enhancing their survival rate and drug resistance.

[0074] 9) Epigenetic Regulation: Key genes BRD7, BAP1, and MLL affect gene expression patterns by regulating DNA methylation and histone modifications, and participate in the occurrence and development of hepatocellular carcinoma.

[0075] 10) Receptor tyrosine kinase (RTK) signaling pathway: Key genes include VEGFR, EGFR, FGFR, HGFR, IGFR, PDGFR, VEGFA, VEGFR2, and IGF2R. These RTKs promote angiogenesis, tumor growth, and drug resistance in hepatocellular carcinoma and are potential therapeutic targets.

[0076] 11) RAS / RAF / MEK / ERK signaling pathway: Abnormal activation of key genes NRAS, HRAS, KRAS, SOS, GRB2, and RAF is common in hepatocellular carcinoma and can promote cancer cell proliferation and survival.

[0077] 12) JAK-STAT signaling pathway: Key genes SALL4, IL6ST, JAK1, SOCS and HSPG2 play a role in regulating cell growth and immune response, and their abnormal activation may accelerate the progression of hepatocellular carcinoma.

[0078] 13) Hedgehog signaling pathway: Abnormal activation of key genes c-MYC, SMO and Gli can enhance the growth and proliferation of liver cancer cells.

[0079] 14) Hippo signaling pathway: Key genes MST1 / 2, LATS1 / 2, YAP, TAZ and TEAD regulate cell proliferation and organ size, among which abnormal activation of YAP / TAZ may promote the growth and progression of hepatocellular carcinoma.

[0080] 15) Apoptosis pathway: Key genes NK1, SSRP1, NF-κB and TP53 regulate cell death signals. Mutations in these genes may lead to hepatocellular carcinoma escaping immune clearance and promoting tumor survival.

[0081] 16) NOTCH signaling pathway: The key gene PSENEN is involved in cell fate determination and proliferation regulation, and its mutation may promote cancer cell survival and proliferation.

[0082] 17) Hypoxia signaling pathway: Key genes HIF1α, HIF2α and c-MYC are activated under hypoxic conditions, promoting tumor adaptation to a hypoxic environment and enhancing angiogenesis.

[0083] 18) Brachyury signaling pathway: The key gene Brachyury may be related to cell differentiation and drug resistance in hepatocellular carcinoma, but its specific mechanism of action is still unclear.

[0084] 19) FGF signaling pathway: Key genes include FGF19, FGFR4, FRS2, GRB2, GAB1, PLCγ, DAG, SPRY, and PKC. This pathway promotes tumor growth, angiogenesis, and drug resistance in hepatocellular carcinoma.

[0085] 20) Cell cycle signaling pathway: Key genes: CDKN2A, RB1, CCNC1, CDK1,2,4,5,6,9, CCND1, and MYC. Together, they form the core network of cell cycle regulation, ensuring that cells replicate and divide at the appropriate time and in the correct order.

[0086] In summary, Figure 1 This study systematically summarizes the core driver genes and key signaling pathways regulated by hepatocellular carcinoma (HCC), covering important pathways such as WNT / β-catenin, RAS / MAPK, PI3K / AKT / mTOR, Hippo, JAK-STAT, Hedgehog, TGF-β, and RTKs. Aberrant activation or dysregulation of these pathways plays a crucial role in the occurrence, progression, and drug resistance of HCC. This research not only deepens our understanding of the molecular mechanisms of HCC but also provides important theoretical basis and potential intervention targets for the development of targeted therapy and personalized treatment strategies.

[0087] 1.2 Gene mutation analysis and organization

[0088] This study systematically compiled and analyzed gene mutation data related to hepatocellular carcinoma from published literature. Next-generation sequencing (NGS) technologies mainly include whole exome sequencing (WES) and whole genome sequencing (WGS). WES focuses on detecting gene mutations in exon regions (i.e., protein-coding regions), while WGS can comprehensively analyze variations across the genome, including exon regions, intron regions, regulatory regions, and structural variations such as single nucleotide variants (SNPs), insertions / deletions (Indels), copy number variations (CNVs), and structural variations (SVs). Driver genes without synonymous mutations refer to functional mutations occurring in the DNA sequence. These mutations are often closely related to genetic diseases and cancers and may lead to significant changes in biological function. This example systematically compiled mutated gene data from published literature, providing important data support for subsequent in-depth analysis.

[0089] 1.2.1 In the first study (DOI: 10.1016 / j.ccell.2023.11.010, PMID38101410), researchers performed a multi-omics integrated analysis on 257 primary tumor regions and 176 metastatic tumor regions from 182 hepatocellular carcinoma patients, covering multiple levels including genomics, transcriptomics, single-cell RNA sequencing (scRNA-seq), spatial transcriptomics, and immunohistochemistry (IHC). The study found that Wnt wild-type metastatic tumors were enriched with immunosuppressive B cells, which induced CD8+ by activating the HLA-E:CD94-NKG2A immune checkpoint pathway. + T cell terminal exhaustion, thereby suppressing anti-tumor immune responses. This study not only reveals the complex mechanisms of hepatocellular carcinoma metastasis and evolution, but also provides multiple potential targets for developing therapeutic strategies against metastatic hepatocellular carcinoma.

[0090] Figure S1(F) in the literature shows common mutated genes in hepatocellular carcinoma in patient cohorts and the TCGA database, including TP53, CTNNB1, ALB, AXIN1, BAP1, KEAP1, NFE2L2, LZTR1, RB1, PIK3CA, RPS6KA3, AZIN1, KRAS, IL6ST, RP1L1, CDKN2A, EEF1A1, ARID2, ARID1A, GPATCH4, ACVR2A, APOB, CREB3L3, NRAS, AHCTF1, and HIST1H1C. This embodiment, based on the TCGA database, calculates the proportion of patients carrying non-synonymous mutations, and the results are shown in Table 2. Furthermore, supplementary information from the literature is also provided. Figure 1 The non-synonymous mutation driver genes PTEN and BRD7 mentioned in (D) further refine the analysis of hepatocellular carcinoma mutation genes.

[0091] 1.2.2 In the second study (DOI: 10.1111 / cas.13582, PMID 29573058), researchers integrated genomic data from 1340 hepatocellular carcinoma cases across different ethnicities, representing the largest hepatocellular carcinoma genomic study to date. This study systematically revealed the core driver gene map of hepatocellular carcinoma, identifying the TP53, TERT, and WNT (CTNNB1) signaling pathways as the three major core driver factors. Furthermore, the study also discovered low-frequency mutation combinations involving multiple cancer signaling pathways, which may play a synergistic role in the occurrence and development of hepatocellular carcinoma. Examples of low-frequency mutated genes include: ALB, AXIN1, ARID1A, ARID2, NFE2L2, CDKN2A, RPS6KA3, RB1, ACVR2A, FGF19, CCND1, APOB, EEF1A1, BRD7, PTEN, PIK3CA, TSC2, BAP1, CDKN2B, KEAP1, KRTAP5-11, CDKN1A, KRAS, RPL22, NBEA, PCF11, TSC1, HNF1A, FGA, and ERRF11. These low-frequency mutated genes may promote the growth, invasion, and drug resistance of liver cancer by affecting mechanisms such as cell cycle, signaling pathway regulation, chromatin remodeling, and metabolic regulation. Furthermore, combinations of mutations in these genes may play a greater role in specific liver cancer subtypes, providing new potential targets for personalized therapy.

[0092] 1.2.3 In the third study (DOI: 10.1158 / 1535-7163.MCT-17-0604, PMID 29483209), researchers systematically studied the frequency of genomic alterations in 26 patients with hepatocellular carcinoma using circulating tumor DNA (ctDNA) analysis. The ctDNA detection technology used in this study can accurately identify single nucleotide variants (SNVs), gene amplifications, fusion mutations, and specific insertion / deletion mutations (indels), covering a range of 54 to 70 genes. The results showed that 23 of the 26 patients (88.5%) had at least one known molecular mutation, and all patients carried at least one potentially targetable mutation. This finding not only reveals the prevalence of genomic alterations in liver cancer patients but also provides important molecular evidence for ctDNA-based precision treatment strategies, possessing potential clinical application value.

[0093] Mutated genes include: TP53, CTNNB1, ARID1A, EGFR, MYC, APC, ATM, CDK6, ERBB2, MET, BRCA1, CCNE1, CDKN2A, FGFR1, FGFR2, KIT, KRAS, NF1, NFE2L2, PIK3CA, RAF1, ALK, AR, BRAF, BRCA2, CCND1, CCND2, CDK4, ESR1, JAK3, MAP2K1, NTRK1, PTEN, RB1, RET, SMAD4, TSC1.

[0094] 1.2.4 The fourth study, based on the GSE273254 dataset from the publicly available Gene Expression Omnibus (GEO) database, focuses on the synergistic clinical significance of TERT promoter mutations (TERTpm) and TP53 gene mutations (TP53m) in hepatocellular carcinoma. The research team collected tumor tissue samples from 50 hepatocellular carcinoma patients undergoing their first radical hepatocellular carcinoma surgery and used next-generation sequencing (NGS) technology to perform a systematic mutation analysis of 1021 tumor-related genes, including TERT and TP53. For example... Figure 2 As shown, the results indicate that among these 50 samples, the TP53 gene mutation occurred in 24 cases (48.0%), and the TERT promoter hotspot mutation C228T occurred in 16 cases (32.0%).

[0095] From this set of data, the following mutated genes were obtained: TP53, TERT, LRP1B, CTNNB1, AXIN1, ARID1A, ATM, TSC2, NOTCH3, APC, KMT2A, ARID2, ROS1, NTRK3, ERBB4, KMT2D, NF1, KMT2C, BCOR, KDR, NOTCH4, ATR, STAT3, NCOR1, EP300, IRF2, HSPD1, KEAP1, TSHR, SYK, PAX5, PDGFRA, BAP1.

[0096] All the mutated genes mentioned in the three studies in 1.2.1 to 1.2.3 above, as well as the results obtained from the analysis of the GSE273254 dataset in 1.2.4 above, have been systematically summarized and integrated in Table 3.

[0097] 1.3 RNA Sequencing Data Analysis

[0098] RNA sequencing (RNA-Seq) is an experimental method based on high-throughput sequencing technology used to comprehensively analyze the expression of transcripts in cells or tissues. Differentially expressed genes (DEGs) are genes whose expression levels change significantly under different experimental conditions or physiological states. This concept has wide applications in transcriptomics, disease mechanism research, and drug development, helping researchers identify key regulatory genes or potential biomarkers. Differentially expressed genes are usually detected using high-throughput sequencing technologies (such as RNA-Seq) or gene chip technology. Based on a negative binomial distribution model, the raw count data can be modeled and analyzed to output a set of genes with significant changes. These changes are typically assessed using p-values ​​or false discovery rates (FDR values) and the magnitude of change (such as the Fold Change value). For example, in cancer research, differentially expressed genes between tumor and normal tissues may be involved in the occurrence, development, or metastasis of tumors.

[0099] DESeq2 is a Bioconductor software package in R for differential expression analysis (DEA) of RNA-seq data. It can directly process raw gene expression counts and, through data normalization and statistical modeling, identify differentially expressed genes (DEGs) under different experimental conditions (such as tumor tissue versus normal tissue).

[0100] 1.3.1 The preferred GSE214846 database is derived from Gene Expression Omnibus (GEO). This study aims to construct a survival prediction model based on immune-related genes (IRGs) to assess the prognosis of hepatocellular carcinoma patients.

[0101] RNA sequencing (RNA-seq) data were obtained from tumors and paired adjacent normal tissues of 65 liver cancer patients from the Affiliated Cancer Hospital of Guangxi Medical University, and gene expression profiling was performed (DOI:10.3389 / fimmu.2022.1023349, PMID36353638).

[0102] Differentially expressed genes were calculated using DESeq2 and identified as follows: DKK1, TERT, SFN, NQO1, CDK1, E2F1, NKD1, CDKN2A, SLC2A1, EF1, HDAC11, TXNRD1, IGF2, E2F3, ACACA, BAK1, SQSTM1, CDK4, SKP2, and PLCG1.

[0103] 1.3.2 The second set of data comes from the Cancer Genome Atlas and Genotype-Tissue Expression (GTEx) database. The data download link is: https: / / toil-xena-hub.s3.us-east- 1. amazonaws.com / download / TcgaTargetGtex_rse m_gene_tpm.gz. In this example, gene expression data from 421 cases of liver hepatocellular carcinoma (LIH C) and 110 cases of normal tissue were extracted from this dataset.

[0104] Differential expression analysis using the DESeq2 software package identified the following significantly differentially expressed genes: REG3A, GPC3, NQO1, AFP, MDK, UBD, AKR1B10, SPP1, IFI27, GLUL, LCN2, IGF2, HULC, CD74, H19, HLA-A, PABPC1, LGALS3BP, GRN, SQSTM1, RP11-386G11.10, ISG15, S100A6, GGH, SPINK1, HLA-F, HLA-B, CTSA, CLIC1, and TM4SF5. Due to the large sample sizes in the TCGA and GTEx databases, the analysis results are highly reliable. Therefore, this example selected the top 30 genes with the most significant differential expression for further analysis. These genes may play an important role in the occurrence, development, and prognosis of hepatocellular carcinoma.

[0105] 1.3.3 The third set of data came from the GSE248562 dataset. This study mainly investigated the effects of environmental monobutyl phthalate (MBP) on the hepatocellular carcinoma cell line HepG2 and the normal hepatocyte cell line L02. RNA-seq sequencing data from eight HepG2 samples and eight L02 samples were selected from the GSE248562 dataset for analysis. Differential expression analysis identified the following significantly differentially expressed genes: APOA1, NR0B2, TTR (transthyretin), SERPINA1, NDRG2, ALDH1A1, CES1, APOC3, FABP1, TF (transferrin), FGA, SERPINA1, APOE, FGG, ALB, HNF4A, ANG (angiogenin), DUSP6, CXCL16, and FN1.

[0106] 1.3.4 The fourth set of data comes from the GSE157905 project in the GEO database. The study focuses on exploring the molecular mechanism and transcriptome characteristics of lenvatinib combined with the EGFR inhibitor gefitinib in the treatment of hepatocellular carcinoma. Four EGFR-high human hepatocellular carcinoma cell lines (SNU449, JHH1, Huh6, SNU182) were selected and treated with DMSO (control), gefitinib, lenvatinib, and a combination of the two for 24 hours, respectively. Subsequently, whole transcriptome RNA sequencing (RNA-seq) was performed (Document DOI: 10.1038 / s41586-021-03741-7, PMID34290403).

[0107] Four samples were selected from SNU449-CT, SNU449-Gefitinib, SNU449-Lenvatinib, and SNU449-Combination, four samples from Huh6, and four samples from SNU182, compared with four samples from JHH1. Differentially expressed genes included: CDH2, SPARC, VIM, MMP2, POSTN, APOE, FN1, FGF2, COL1A1, COL1A2, LOXL2, AKT3, LOX, NRP1, TGFB1, SERPINE1, WNT5A, CAV1, CMBL, and IRS1.

[0108] 1.3.5 The fifth set of data came from dataset GSE117623 in the GEO database. This study focused on the transcriptomic characteristics of circulating epithelial cells (CECs) in the peripheral blood of patients with chronic liver disease (CLD) and hepatocellular carcinoma (DOI: 10.1053 / j.gastro.2018.09.020, PMID30218669). Fifty-two hepatocellular carcinoma samples and 65 normal liver tissue samples were analyzed, and the following genes were screened through differential expression analysis: CA1, OSBP2, TRIM58, MXI1, PRDX2, MAP2K3, SELENBP1, ADIPOR1, BLVRB, HBD, RBM38, BAG1, BNIP3L, FOXO3, BSG, CTSB, BCL2L1, DUSP1, EIF5, and HK1.

[0109] 1.3.6 The Cancer Cell Line Encyclopedia Database (CCLE) is a comprehensive and authoritative large-scale cancer cell line dataset, covering cell lines of various cancer types and their detailed molecular characteristics. Twenty-eight hepatocellular carcinoma (HCC) cell lines were extracted from the CCLE database, providing important experimental materials for studying the molecular mechanisms of HCC. To explore the role of epithelial-mesenchymal transition (EMT) in HCC, these cell lines were ranked according to the expression level of the vimentin gene (VIM). VIM is one of the key markers of EMT, and its high expression is closely related to the EMT process, which plays a crucial role in tumor cell migration, invasion, and metastasis. Based on VIM expression levels, the three cell lines with the highest and lowest VIM expression levels were further selected for comparative analysis: the high VIM expression group included JHH2, SNU449, and SKHEP1, while the low VIM expression group consisted of JHH6, HEPG2, and JHH5, as shown in Table 4. This analysis provides important experimental evidence for further research on the role of EMT in hepatocellular carcinoma.

[0110] Differential expression analysis identified the following differentially expressed genes: LOXL, CXCL1, EFEMP1, VIM, MYL9, BGN, TNC, IL1B, PMP22, COL1A1, CCL2, SRI, ETS1, FSTL1, SRPX, ITGA5, CXCL8, VCAN, VLDLR, and PMEPA1.

[0111] 1.3.7 The seventh set of data comes from dataset GSE186191 in the GEO database (DOI: 10.1002 / hep4.1928, PMID35238496). Transcriptome sequencing (RNA-seq) was performed on lenvatinib-resistant liver cancer cell models (Hep3B-LR, Huh7-LR) and their corresponding parental cells (Hep3B-P, Huh7-P). The DESeq2 algorithm was used to screen for significantly differentially expressed genes, as shown in Table 5: CCNG2, RRAGD, ME1, LDHA, PPIA, ME2, SERPINE1, PPP1R15A, DSC2, ZNF292, KDELR3, NAGK, SLC16A3, GMPPA, LGALS3, HMMR, JMJD6, MMP14, IER3, MMP15, TP53, and HDAC2.

[0112] Ultimately, this embodiment yielded 619 potential gene knockout targets, as shown in Table 7.

[0113] Example 2

[0114] This example demonstrates data filtering.

[0115] The specific details of this embodiment are as follows:

[0116] In this embodiment, all genes obtained in Example 1 are summarized and then screened according to the following algorithm:

[0117] 2.1 Risk Ratio

[0118] The hazard ratio (HR) is a commonly used statistical indicator in survival analysis, used to measure the impact of a gene expression level or clinical characteristic on patient survival time. The HR value can intuitively reflect the patient's survival risk: when HR>1, it indicates that patients with high expression of the gene have an increased survival risk; when HR<1, it indicates a decreased survival risk.

[0119] In this embodiment, HR analysis was performed using hepatocellular carcinoma (LIHC) data from the TCGA database, and the calculation was performed using the Cox Proportional-Hazards Model.

[0120] First, download the clinical data of hepatocellular carcinoma patients from the TCGA database (file link: https: / / gdc- hub.s3.us-east-1.amaThe file (zonaws.com / download / TCGA-LIHC.clinical.tsv.gz) contains patient survival information. The survival status (status) is either 0 or 1: 0 indicates the event did not occur (patient survives), and 1 indicates the event occurred (patient dies). Gene expression data for each hepatocellular carcinoma patient was also downloaded (file link: [link missing]). https: / / gdc-hub.s3.us-east-1.amazonaws.com / download / TCG (A-LIHC.star_tpm.tsv.gz). Clinical data and gene expression data are integrated using Perl scripts for subsequent analysis.

[0121] Next, Cox regression analysis was performed using the `survival` and `survminer` packages in R to calculate the HR value for each gene. Specifically, the `exp(coef)` value, or HR value, was extracted from the Cox regression model to assess the impact of gene expression on patient survival risk. Finally, the top 40 genes with the highest HR values ​​were selected from the analysis results, as shown in Table 6. These genes may be closely related to the survival risk of hepatocellular carcinoma patients, providing important candidate targets for further research.

[0122] 2.2 Survival Analysis

[0123] The survival analysis employed the Kaplan-Meier tool, a powerful statistical tool capable of assessing the correlation between the expression levels of various molecules (such as mRNA, miRNA, protein, and DNA) and patient survival. This tool visually demonstrates the impact of different gene expression levels on patient prognosis by plotting survival curves. This example focuses on two key survival indicators: overall survival (OS) and relapse-free survival (RFS). Overall survival (OS) refers to the time from patient diagnosis or initiation of treatment to death from any cause; it is a core indicator for measuring treatment effectiveness and patient prognosis, comprehensively reflecting the patient's survival status. The length of OS directly reflects the degree of threat the disease poses to the patient's life and the effectiveness of treatment. Relapse-free survival (RFS) refers to the time from successful treatment (e.g., complete surgical resection of the tumor) to disease recurrence or patient death (regardless of cause); RFS is primarily used to assess whether treatment can effectively control disease recurrence, and it is particularly significant in cancer research; it helps researchers understand the disease control effect after treatment and the long-term quality of life of patients.

[0124] To analyze the correlation between gene expression and patient survival, this embodiment used Kaplan-Meier survival curves to assess overall survival (OS) and response time (RFS), respectively. Specific analysis was performed using the online tool KMplot(…). https: / / www.kmplot.com / analysis / index.php? p=service&c The analysis was completed using the parameter `ancer = liver_rnaseq`. During the analysis, the following parameters were set: Splitpatients were automatically selected to differentiate between high-expression and low-expression groups; Analysis subjects: All target genes were input into the website, and their expression levels were assessed in relation to overall survival (OS) and response time (RFS). After analysis, the OS and RFS results for each gene were recorded, focusing on the following conditions: High expression and short survival: Whether patients with high gene expression exhibited shorter OS or RFS; Statistical significance: The p-value for each gene was recorded to determine whether the correlation between its expression level and survival rate was statistically significant. Through the above analysis, genes closely related to the survival of hepatocellular carcinoma patients were identified, providing important evidence for further research on their biological functions and clinical significance. According to the analysis results, the 49 genes with statistically significant differences in overall survival and recurrence-free survival are: ACA CA, ATR, AXIN1, BRCA1, BSG, CCNE1, CDK1, CDK2, CDK4, CDK5, CDKN2A, CDKN2D, CLIC1, CTAG1B, E2F1, E2F2, E2F3, EIF4E, EWSR1, EZH2, FUBP1, GPATCH4, GRB2, GRN, HIST1H3C, HMMR, HRAS, HSPD1, ME2, MMP1, MYB, OSBP2, PAX8, PLCB1, PPIA, POLD 1, SFN, SMARCD1, SNRPE, SPP1, SQSTM1, SRI, SSRP1, STC1, TERT, TNNT1, TSC1, TXN RD1, and VEGFA. These can all serve as potential targets for hepatocellular carcinoma.

[0125] 2.3 Expression levels at different tumor stages (stages I-IV)

[0126] Calculate gene expression levels at different tumor stages (I-IV). Download clinical data of hepatocellular carcinoma patients from TCGA, including clinical information on tumor stage. File link: https: / / gdc-hub.s3.us-east- 1. amazonaws.com / download / TCGA-LIHC .clinical.tsv.gz; Download RNA sequencing expression data from the same project, including the expression level of each gene in each patient (in Transcripts Per Million, TPM value), file link: https: / / gdc-hub.s3.us-east-1.amazonaws.comThe file is located at / download / TCGA-LIHC.star_tpm.tsv.gz. It uses a Perl script to integrate clinical data and RNAseq expression data, allowing for further analysis of gene expression differences across different tumor stages (I-IV), as well as between early (I-II) and late (III-IV) stages.

[0127] 2.4 Analysis based on UALCAN

[0128] Based on UALCAN ( https: / / ualcan.path.uab.edu / analysis.html This embodiment of the study analyzed the expression levels of genes in hepatocellular carcinoma (HCC) using the following methods: Sample Types: The expression levels of genes in normal liver tissue and tumor tissue were compared to assess their expression changes in HCC and preliminarily determine their activation or inhibition during the carcinogenesis process. Individual Cancer Stages: The expression differences of each gene in different tumor stages (I, II, III, IV) were analyzed to assess their relationship with tumor progression. Tumor Grade: The expression differences of each gene in different tumor grades (1, 2, 3, 4) were analyzed to assess their correlation with tumor differentiation and malignancy. Nodal Metastasis Status: The expression differences of each gene in patients with and without lymph node metastasis were analyzed to study their regulatory role in tumor metastasis. These analyses provide important evidence for revealing the functions and clinical significance of genes in HCC.

[0129] 2.5 Assessment of Research Activity Based on the Number of Literature Quantities

[0130] The number of research articles related to hepatocellular carcinoma (HCC) for the gene was retrieved from the NCBI database and found in PubMed. PubMed literature support assessment: All candidate genes were searched in the PubMed database using the keywords "Gene Name" + "HCC," and the number of relevant articles was counted to determine the research activity of the gene in the field of liver cancer.

[0131] Based on the results in sections 2.3 to 2.5 above, the following 46 genes can be considered as candidate gene knockout targets for precision treatment of hepatocellular carcinoma: AKT1, ATM, ATR, AXIN1, AXIN2, BAK1, BRCA1, BSG, CCNE1, CDK1, CDK2, CDK5, CDK6, CDKN2B, DNMT3A, E2F3, EIF4E, EPCAM, EZH2, FGFR1, FOXO3, FZD7, GLI1, HDAC2, HMMR, HMOX1, HRAS, IDH1, IGF1R, IL1B, KIT, KRAS, MUC1, MYB, NOTCH1, PDK1, POSTN, PPARG, RB1, SALL4, SERPINE1, SMAD2, SRI, TGFBR1, TSC1, and YAP1. A schematic diagram illustrating the relevant data for five candidate gene knockout targets for hepatocellular carcinoma is shown below. Figure 3 As shown.

[0132] Example 3

[0133] This example demonstrates target validation.

[0134] The specific details of this embodiment are as follows:

[0135] This embodiment analyzes the expression levels of each gene in different hepatocellular carcinoma (HCC) cancer cell lines based on the candidate gene knockout target library selected in Example 2 for precision treatment of HCC. Data was sourced from the CCLE database. Results show that each gene knockout target is expressed in different HCC cancer cell lines, with varying expression patterns. Some genes exhibit significant high or low expression trends in specific cell lines. These validation results serve as the expression background of each gene knockout target in different cell lines, providing important evidence and clear direction for subsequent functional experiments (e.g., studying the function of genes in HCC and their potential as therapeutic targets). Researchers can use this information to select appropriate cell lines for functional validation experiments, further explore the role of genes in the occurrence, development, and treatment of HCC, and select suitable models for functional knockout, overexpression, and drug intervention experiments, thereby accelerating gene function validation and drug sensitivity assessment. Furthermore, these validation results can provide a basic reference for subsequent CRISPR screening, cross-validation of drug sensitivity databases, and the construction of personalized treatment models, demonstrating the bridging role of this invention between basic research and clinical translation.

[0136] The following are the tables involved in the above embodiments.

[0137] Table 1. Signaling pathways and genes related to hepatocellular carcinoma and their corresponding PMIDs retrieved from the PubMed database in Example 1.

[0138]

[0139]

[0140]

[0141] Table 2. Proportion of hepatocellular carcinoma patients carrying nonsynonymous mutations in Example 1 based on the TCGA database.

[0142] Gene TCGA mutation percentage (%) TP53 31 CTNNB1 27 ALB 13 AXIN1 8 BAP1 5 KEAP1 5 NFE2L2 3 LZTR1 3 RB1 4 PIK3CA 4 RPS6KA3 4 AZIN1 3 KRAS 1 IL6ST 3 RP1L1 3 CDKN2A 2 EEF1A1 3 ARID2 5 ARID1A 7 GPATCH4 2 ACVR2A 3 APOB 11 CREB3L3 1 NRAS 1 AHCTF1 4 HIST1H1C 2

[0143] Table 3. Systematic compilation of nonsynonymous mutant gene data related to hepatocellular carcinoma in Example 1.

[0144]

[0145] Table 4. Analysis of VIM gene expression levels in hepatocellular carcinoma cell lines in Example 1 using the CCLE database.

[0146]

[0147] Table 5. Transcriptome comparison of lenvatinib-resistant hepatocellular carcinoma cell lines (Hep3B-LR, Huh7-LR) and their parental sensitive lines in the GSE186191 dataset in Example 1. The log2 fold change of the top 22 differentially expressed candidate genes was extracted.

[0148] log2BaseMean (log-mean expression level)

[0149]

[0150] Table 6. Example 2 identified the top 40 genes with the highest Hazard Ratio (HR).

[0151]

[0152]

[0153] Table 7 shows that a total of 619 potential gene knockout targets were obtained in Example 1.

[0154]

[0155]

[0156] In addition to the embodiments described above, the present invention may have other implementations. All technical solutions formed by equivalent substitution or equivalent transformation fall within the protection scope claimed by the present invention.

[0157] Based on the above specific embodiments, the overall flow diagram of the present invention is as follows: Figure 4 As shown.

[0158] This invention comprehensively reveals the molecular mechanisms of liver cancer through the integrated analysis of multiple datasets, capturing complex biological processes that cannot be reflected by single-data sets. For example, genomic data can identify driver mutations, transcriptomic data can reveal gene expression regulatory networks, proteomic data can reflect protein function and interactions, and metabolomic data can reveal the characteristics of tumor metabolic reprogramming. This multi-dimensional data integration not only improves the comprehensiveness and accuracy of target screening but also discovers new potential therapeutic targets and biomarkers.

[0159] In the design of the algorithm flow, this invention first preprocesses and standardizes multi-omics data to eliminate batch effects and technical biases. Then, key features are extracted through feature selection and dimensionality reduction techniques, and target prediction is performed using bioinformatics algorithms. Furthermore, clinical data (such as patient survival) is incorporated to construct a predictive model to assess the clinical relevance of the targets, ensuring high reliability and reproducibility of the selected targets. This not only significantly improves the accuracy and efficiency of target screening but also provides important theoretical basis and technical support for precision treatment and drug development in hepatocellular carcinoma. For example, based on the screened key genes, targeted gene knockout experiments (such as CRISPR-Cas9) can be designed to verify their function and mechanism of action in hepatocellular carcinoma. Simultaneously, these targets can be used to develop new targeted drugs or optimize the efficacy of existing drugs, accelerating the research and development process of liver cancer drugs. In addition, the target library can provide clinicians with personalized treatment recommendations, selecting the most suitable targeted drugs or immunotherapy regimens based on the patient's gene mutation profile and expression characteristics.

[0160] From both scientific research and clinical perspectives, the technical solution of this invention has significant application value. In scientific research, it provides a tool for liver cancer research, helping to obtain key gene information, verify functions, and analyze mechanisms. Clinically, it supports precision medicine, assists in the development of treatment plans, and improves patient prognosis. For example, by analyzing tumor samples to identify driver genes or drug resistance mechanisms, treatment strategies can be optimized.

[0161] The high incidence, high mortality, and treatment challenges of hepatocellular carcinoma highlight the urgency of research. Gene knockout technology, as an important tool in functional genomics, plays a crucial role in liver cancer research, drug development, and clinical treatment. Through multidisciplinary collaboration, combining multi-omics and bioinformatics technologies, the technical solution of this invention holds the promise of promoting precision treatment for liver cancer and improving patient prognosis. Its application prospects include: Precision treatment: providing personalized targeted therapies; Drug development: accelerating new drug discovery and drug retargeting; Mechanism research: elucidating the mechanisms of liver cancer occurrence, development, and metastasis; Biomarker discovery: identifying biomarkers related to diagnosis, prognosis, and treatment. With technological advancements, the technical solution of this invention is expected to bring breakthrough progress to liver cancer research and treatment.

Claims

1. A method for screening hepatocellular carcinoma gene knockout target libraries based on multi-omics, characterized in that, Includes the following steps: Step 1: Data Collection and Integration S1. Search public databases to collect reported gene information data related to hepatocellular carcinoma, and after systematic analysis, draw a map of the core driver genes of hepatocellular carcinoma and the key signaling pathways they regulate. S2. Systematically organize and analyze gene mutation data related to hepatocellular carcinoma in published literature, including: obtaining gene mutation data in protein-coding regions by organizing and analyzing whole-exome sequencing results; obtaining genome-wide variation data, including variation data in exon regions, intron regions, and regulatory regions, by organizing and analyzing whole-genome sequencing results. S3. Based on RNA sequencing data, the DESeq2 software tool in R language was used to analyze and identify differentially expressed genes (DEGs) between tumor and normal tissues. The second step is data screening: After summarizing all the genes obtained in the first step, risk ratio analysis, survival analysis, different tumor stage analysis, UALCAN-based analysis, and research activity assessment analysis based on the number of literatures are performed. Then, all the analysis results are summarized to obtain a candidate gene knockout target library for precision treatment of hepatocellular carcinoma. Step 3, Target Validation: For the candidate gene knockout target library for precision treatment of hepatocellular carcinoma obtained in Step 2, the expression level and expression pattern of each gene in different hepatocellular carcinoma cell lines are analyzed, and genes that are not expressed in all hepatocellular carcinoma cell lines are removed; finally, a gene knockout target library for precision treatment of hepatocellular carcinoma is obtained.

2. The method for screening a hepatocellular carcinoma gene knockout target library based on multi-omics as described in claim 1, characterized in that, In step S1, the public database includes the PubMed database of the National Center for Biotechnology Information (NCBI). The gene information data related to hepatocellular carcinoma includes: genes, signaling pathways, mutation profiles, expression profiles, epigenetic modifications, protein-protein interaction networks, and metabolic pathway data; the atlas includes: cell cycle regulation, telomere maintenance, WNT / β-catenin signaling pathway, chromatin remodeling, hepatocyte differentiation, RAS / MAPK / PI3K / AKT / mTOR signaling pathway, TGF-β signaling pathway, oxidative stress resistance, epigenetic regulation, receptor tyrosine kinase signaling pathway, RAS / RAF / MEK / ERK signaling pathway, JAK-STAT signaling pathway, Hedgehog signaling pathway, Hippo signaling pathway, apoptosis pathway, NOTCH signaling pathway, hypoxia signaling pathway, Brachyury signaling pathway, and FGF signaling pathway.

3. The method for screening a hepatocellular carcinoma gene knockout target library based on multi-omics as described in claim 1, characterized in that, In step S2, the variation data specifically includes: single nucleotide variations, insertions / deletions, copy number variations, and structural variations.

4. The method for screening a hepatocellular carcinoma gene knockout target library based on multi-omics as described in claim 1, characterized in that, The specific process of step S2 includes: S2-1. Existing literature conducted a multi-omics integrated analysis of 257 primary tumor regions and 176 metastatic tumor regions from 182 hepatocellular carcinoma patients. Based on this existing literature, a systematic review and analysis were conducted, and the following hepatocellular carcinoma mutated genes were found to be commonly found in the Cancer Genome Atlas (TCGA): TP53, CTNNB1, ALB, AXIN1, BAP1, KEAP1, NFE2L2, LZTR1, RB1, PIK3CA, RPS6KA3, AZIN1, KRAS, IL6ST, RP1L1, CDKN2A, EEF1A1, ARID2, ARID1A, GPATCH4, ACVR2A, APOB, CREB3L3, NRAS, AHCTF1, HIST1H1C, PTEN, and BRD7. S2-2. Existing literature was integrated to collect genomic data from 1340 hepatocellular carcinoma cases from different ethnic groups. Based on this existing literature, a systematic review and analysis were conducted, and the TP53, TERT, and WNT (CTNNB1) signaling pathways were identified as the three core driving factors. Low-frequency mutated genes involved in multiple cancer signaling pathways were identified as including: ALB, AXIN1, ARID1A, ARID2, NFE2L2, CDKN2A, RPS6KA3, RB1, ACVR2A, FGF19, CCND1, APOB, EEF1A1, BRD7, PTEN, PIK3CA, TSC2, BAP1, CDKN2B, KEAP1, KRTAP5-11, CDKN1A, KRAS, RPL22, NBEA, PCF11, TSC1, HNF1A, FGA, and ERRF11. S2-3. Existing literature has systematically studied the frequency of genomic alterations in 26 patients with hepatocellular carcinoma through circulating tumor DNA analysis. Based on the systematic review and analysis of this existing literature, the mutated genes identified include: TP53, CTNNB1, ARID1A, EGFR, MYC, APC, ATM, CDK6, ERBB2, MET, BRCA1, CCNE1, CDKN2A, FGFR1, FGFR2, KIT, KRAS, NF1, NFE2L2, PIK3CA, RAF1, ALK, AR, BRAF, BRCA2, CCND1, CCND2, CDK4, ESR1, JAK3, MAP2K1, NTRK1, PTEN, RB1, RET, SMAD4, and TSC1. S2-4. Analysis of existing literature: The GSE273254 dataset in the public database GEO was analyzed. Based on this existing literature, a systematic organization and analysis were conducted, and the mutated genes identified included: TP53, TERT, LRP1B, CTNNB1, AXIN1, ARID1A, ATM, TSC2, NOTCH3, APC, KMT2A, ARID2, ROS1, NTRK3, ERBB4, KMT2D, NF1, KMT2C, BCOR, KDR, NOTCH4, ATR, STAT3, NCOR1, EP300, IRF2, HSPD1, KEAP1, TSHR, SYK, PAX5, PDGFRA, and BAP1. S2-5. Summarize the mutated genes obtained from the above steps to obtain gene mutation data related to hepatocellular carcinoma.

5. The method for screening a hepatocellular carcinoma gene knockout target library based on multi-omics as described in claim 1, characterized in that, The specific process of step S3 includes: S3-1. Existing research has constructed survival prediction models for immune-related genes based on the GSE214846 database. One study selected 65 liver cancer patients and their paired adjacent normal tissues for RNA sequencing and gene expression profiling analysis. Based on the RNA sequencing data from this existing study, differentially expressed genes were first screened using the DESeq2 software, including: DKK1, TERT, SFN, NQO1, CDK1, E2F1, NKD1, CDKN2A, SLC2A1, EF1, HDAC11, TXNRD1, IGF2, E2F3, ACACA, BAK1, SQSTM1, CDK4, SKP2, and PLCG1. S3-2. Utilizing existing cancer genome maps and the GTEx genotype tissue expression database, gene expression data from 421 hepatocellular carcinoma cases and 110 normal tissue cases were extracted from this database. Differentially expressed genes were first screened using the DESeq2 software, including: REG3A, GPC3, NQO1, AFP, MDK, UBD, AKR1B10, SPP1, IFI27, GLUL, LCN2, IGF2, HUL C, CD74, H19, HLA-A, PABPC1, LGALS3BP, GRN, SQSTM1, RP11-386G11.10, ISG15, S100A6, GGH, SPINK1, HLA-F, HLA-B, CTSA, CLIC1, and TM4SF5. S3-3. Using the existing GSE248562 dataset, RNA sequencing data from eight HepG2 samples and eight L02 samples were selected from this dataset. The differentially expressed genes were first screened using the DESeq2 software, including: APOA1, NR0B2, TTR (transferrin), SERPINA1, NDRG2, ALDH1A1, CES1, APOC3, FABP1, TF (transferrin), FGA, SERPINA1, APOE, FGG, ALB, HNF4A, ANG (angiogenin), DUSP6, CXCL16 and FN1. S3-4. Utilize the existing GSE157905 project; select data from four samples (SNU449-CT, SNU449-Gefitinib, SNU449-Lenvatinib, SNU449-Combination, Huh6), four samples (SNU182), and four samples (JHH1) from this project. First, use the software DESeq2 to analyze and screen differentially expressed genes, including: CDH2, SPARC, VIM, MMP2, POSTN, APOE, FN1, FGF2, COL1A1, COL1A2, LOXL2, AKT3, LOX, NRP1, TGFB1, SERPINE1, WNT5A, CAV1, CMBL, and IRS1. S3-5. Using the existing GSE117623 dataset, data from 52 hepatocellular carcinoma cases and 65 normal liver tissue samples were selected from this project. The differentially expressed genes were first screened using the DESeq2 software, including: CA1, OSBP2, TRIM58, MXI1, PRDX2, MAP2K3, SELENBP1, ADIPOR1, BLVRB, HBD, RBM38, BAG1, BNIP3L, FOXO3, BSG, CTSB, BCL2L1, DUSP1, EIF5, and HK1. S3-6. Using the existing Cancer Cell Line Encyclopedia (CCLE) database, we screened out the VIM high-expression groups JHH2, SNU449, and SKHEP1, and the VIM low-expression groups JHH6, HEPG2, and JHH5. Based on their data, we first used the DES eq2 software to analyze and screen out differentially expressed genes, including: LOXL, CXCL1, EFEMP1, VIM, MYL9, BGN, TNC, IL1B, PMP22, COL1A1, CCL2, SRI, ETS1, FSTL1, SRPX, ITGA5, CXCL8, VCAN, VLDLR, and PMEPA1. S3-7. Using the existing GSE186191 dataset, RNA sequencing data of lenvatinib-resistant liver cancer cell models Hep3B-LR and Huh7-LR, and corresponding parental cells Hep3B-P and Huh7-P, were selected from this dataset. Differentially expressed genes were first screened using the DESeq2 software, including: CCNG2, RRAGD, ME1, LDHA, PPIA, ME2, SERPINE1, PPP1R15A, DSC2, ZNF292, KDELR3, NAGK, SLC16A3, GMPPA, LGALS3, HMMR, JMJD6, MMP14, IER3, MMP15, TP53, and HDAC2.

6. The method for screening a hepatocellular carcinoma gene knockout target library based on multi-omics as described in claim 1, characterized in that, The second step, the specific process of the risk ratio analysis, includes: Clinical data of hepatocellular carcinoma (HCC) patients and gene expression data for each HCC patient were obtained from the Cancer Genome Atlas (TCGA). Cox regression analysis was performed using the R software tools survival and survminer to calculate the hazard ratio (HR) of each gene. Genes with the highest HR values ​​were selected from the analysis results, including: TBXT, WNT8A, CSNK1A1L, TERC, PIK3R2, GSTT4, ATR, WNT16, NTRK1, ALK, EPB42, RP1L1, BDNF, ATM, TSHR, APC2, SEM1, PRKCG, SNCA, TRIM58, SF3B1, KMT2D, MYB, PPIA, SMARCD1, COL1A2, AKT2, SMAD2, POLK, FUBP1, ZNF292, HSPD1, AXIN1, CNGA3, PCF11, HULC, GSTA3, MK RN1, FUS, and CDK2.

7. The method for screening a hepatocellular carcinoma gene knockout target library based on multi-omics as described in claim 6, characterized in that, In the second step, the specific process of the survival analysis includes: Kaplan-Meier survival curves were plotted to assess the correlation between gene expression levels and overall survival (OS) and relapse-free survival (RFS). During the analysis, patients were automatically categorized into high-expression and low-expression groups by selecting the optimal cutoff value. After analysis, OS and RFS results for each gene were recorded, with particular attention paid to genes with high expression and short OS or RFS. Statistical significance was determined by recording the p-value for each gene to determine whether the correlation between its expression level and OS and RFS was statistically significant. Genes identified as statistically significant included: ACACA, ATR, AXIN1, BRCA1, BSG, CCNE1, and C. DK1, CDK2, CDK4, CDK5, CDKN2A, CDKN2D, CLIC1, CTAG1B, E2F1, E2F2, E2F3, EIF4E, EWSR1, EZH2, FUBP1, GPATCH4, GRB2, GRN, HIST1H3C, HMMR, HRAS, HSPD1, ME2, MMP1, MYB, OSBP2, PAX8, PLCB1, PPIA, POLD1, SFN, SMARCD1, SNRPE, SPP1, SQSTM1, SRI, SSRP1, STC1, TERT, TNNT1, TSC1, TXNRD1, VEGFA.

8. The method for screening a hepatocellular carcinoma gene knockout target library based on multi-omics as described in claim 7, characterized in that, The second step involves the following specific procedures for analyzing different tumor stages: Clinical data of hepatocellular carcinoma patients were obtained from the Cancer Genome Atlas (TCGA), along with RNA sequencing expression data containing the expression levels of each gene in each patient. Then, the differences in the expression levels of each gene in different tumor stages I-IV, as well as in early stages I-II and late stages III-IV, were analyzed. The specific process of the UALCAN-based analysis includes: A systematic analysis of gene expression in hepatocellular carcinoma was conducted, including: Sample type analysis: comparing the expression differences of each gene in normal liver tissue and hepatocellular carcinoma tissue to preliminarily determine their activation or inhibition during carcinogenesis; Cancer staging analysis: analyzing the expression differences of each gene in different tumor stages I-IV and assessing their correlation with tumor progression; Tumor grading analysis: analyzing the expression differences of each gene in different tumor grades 1-4 and assessing their correlation with tumor differentiation and malignancy; Lymph node metastasis status analysis: analyzing the expression differences of each gene in patients with and without lymph node metastasis to study their regulatory role in tumor metastasis. The specific process of the research activity assessment and analysis based on the number of documents includes: Using the PubMed database of the National Center for Biotechnology Information (NCBI), we retrieved literature on each gene in the field of hepatocellular carcinoma research and counted the number of related literatures to determine the research activity of the gene in the field of liver cancer. Based on the combined results of the three analyses above, the identified genes include: AKT1, ATM, ATR, AXIN1, AXIN2, BAK1, BRCA1, BSG, CCNE1, CDK1, CDK2, CDK5, CDK6, CDKN2B, DNMT3A, E2F3, EIF4E, EPC AM, EZH2, FGFR1, FOXO3, FZD7, GLI1, HDAC2, HMMR, HMOX1, HRAS, IDH1, IGF1R, IL1B, KIT, KRAS, MUC1, MYB, NOTCH1, PDK1, POSTN, PPARG, RB1, SALL4, SERPINE1, SMAD2, SRI, TGFBR1, TSC1, and YAP1.

9. The method for screening a hepatocellular carcinoma gene knockout target library based on multi-omics as described in claim 1, characterized in that, In the third step, relevant data from the CCLE (Encyclopedia of Cancer Cell Lines) database were used for analysis.

10. A gene knockout target library for precision treatment of hepatocellular carcinoma obtained by the screening method according to any one of claims 1 to 9.

Images