Gp96 tumor neoantigen prediction method and application

Abstract

The application discloses a gp96 tumor neoantigen prediction method and application thereof, and relates to the technical field of tumor neoantigen prediction. The method comprises the following steps: S1, isolation and purification of gp96-tumor polypeptide complexes; S2, dissociation and purification of polypeptides; S3, mass spectrometric identification and sequence analysis of mutant peptides; S4, MHC affinity prediction; and S5, immunogenicity verification. The application takes gp96 as a natural biological concentrator, the captured polypeptides include products in a natural antigen processing path of tumor cells, and antigen peptide segments with potential immunological significance are enriched. In combination with MHC affinity prediction and immunogenicity function verification, the true positive rate of neoantigen screening is significantly improved. In a verification experiment on colon cancer samples, among 7 candidate neoantigens obtained through prediction, 4 can significantly activate autologous T cells of a patient to secrete IFN-gamma, and show a high immunogenicity verification positive rate.

Description

Technical Field

[0001] This invention relates to the field of tumor neoantigen prediction technology, and in particular to a method and application for predicting tumor neoantigens based on gp96. Background Technology

[0002] Tumor neoantigens are abnormal protein fragments produced by gene mutations in tumor cells and specifically recognized by the host immune system. They are expressed only in tumor cells and do not participate in central immune tolerance, making them ideal targets for inducing anti-tumor immune responses. They have significant application value in areas such as personalized cancer vaccines and adoptive T-cell therapy. Accurate prediction and screening of neoantigens are the core prerequisites for developing related immunotherapies.

[0003] gp96 (also known as glucose-regulated protein 94, GRP94) is an important member of the heat shock protein family in the endoplasmic reticulum (ER). Studies have found that gp96, as a molecular chaperone protein, plays a crucial role in the MHC class I antigen presentation pathway. It can specifically bind to peptides that have undergone preliminary processing in the ER and deliver them to newly generated MHC class I molecules, thereby participating in the loading and quality regulation of antigenic peptides. More importantly, in tumor cells, due to genomic instability and high-frequency mutations, the peptide library bound to gp96 is enriched with a large number of peptides derived from tumor-specific mutant proteins, i.e., potential tumor neoantigens. Therefore, gp96 can be considered a natural "bioaccumulator" located at a critical node in antigen processing, and the peptides it binds to directly reflect the antigen profile transported by tumor cells to the ER that has the potential to be presented by MHC class I molecules.

[0004] Based on this, gp96 is used as a tool to directly isolate naturally bound peptide libraries from tumor tissues or cells, and high-throughput mass spectrometry is used for identification, providing a promising "bottom-up" technical approach for discovering real and reliable neoantigens. This method directly addresses the functional antigen processing and presentation process, and the obtained peptides are products of natural intracellular selection, greatly increasing the probability of discovering neoantigens that are actually presented and may have immunogenicity.

[0005] The current mainstream technical routes for predicting tumor neoantigens are bioinformatics prediction methods based on genome / transcriptome sequencing (hereinafter referred to as "sequencing methods") and immunopeptidomics methods based on MHC-peptide complex enrichment (hereinafter referred to as "MHC enrichment methods"). Both types of methods have significant shortcomings in process design and performance.

[0006] First, bioinformatics prediction methods based on genome / transcriptome sequencing (sequencing methods) identify somatic mutations by performing whole-genome sequencing (WGS) or whole-exome sequencing (WES) on tumor tissue and matched normal tissue. Then, bioinformatics algorithms such as NetMHC, NetMHCpan, and MHCflurry are used to predict the binding affinity between the peptide encoded by the mutation and the MHC molecule corresponding to the patient's specific HLA subtype. Based on the prediction score (such as IC50 value, percentile ranking), candidate neoantigens with high affinity are screened. Although this method can systematically screen potential neoantigens, bioinformatics models cannot perfectly simulate the complex antigen processing and presentation process in vivo, resulting in problems such as limited prediction accuracy, large workload for experimental verification, and high cost.

[0007] Secondly, the MHC-peptide complex enrichment-based immunopeptidomics method (MHC enrichment method) enriches MHC molecules (class I or II) from tumor cells or tissues (or antigen-presenting cells) through immunoprecipitation, obtains the naturally presenting peptides bound to them through acid elution, and then identifies them using LC-MS / MS to obtain the true immunopeptide genome. This method can directly obtain information on peptides presented on the surface of tumor cells, but it is heavily dependent on the patient's HLA typing and limited by the high polymorphism of MHC molecules and the downregulation of MHC molecules by tumor cells. Because different patients carry vastly different MHC alleles, it is necessary to construct specific antibodies against specific MHC alleles, and the extraction efficiency is extremely low in tumor tissues with low MHC expression, severely limiting the versatility and sensitivity of this method. Therefore, this invention provides a gp96-based tumor neoantigen prediction method and its application. Summary of the Invention

[0008] The main objective of this invention is to provide a gp96-based tumor neoantigen prediction method and its application that features high prediction accuracy and efficient experimental procedures.

[0009] To achieve the above objectives, the technical solution adopted by the present invention is as follows: This invention provides a method for predicting tumor neoantigens based on gp96, which includes the following steps: Isolation and purification of S1.gp96-tumor polypeptide complex: Total protein was extracted from patient tumor tissue samples, purified by two steps of ConA affinity chromatography and HiTrap Q ion exchange, concentrated and sterilized, and gp96-tumor polypeptide complex was specifically isolated and purified. S2. Dissociation and purification of peptides: The gp96-tumor peptide complex was incubated under acidic conditions containing trifluoroacetic acid to dissociate the peptides from gp96. The dissociated peptide mixture was desalted and purified into low molecular weight peptides by C18 solid phase extraction column. S3. Mass Spectrometry Identification and Sequence Analysis of Mutant Peptides: The purified peptides were analyzed by liquid chromatography-tandem mass spectrometry (LC-MS / MS). The mass spectrometry data were analyzed using PEAKS Studio software to identify the amino acid sequences of the peptides. The identified peptide sequences were compared with the dbSNP public single nucleotide polymorphism database to exclude peptides with known germline polymorphisms, thus obtaining candidate peptides. Peripheral blood DNA was extracted from the patient for sequencing verification. If the gene sequence corresponding to the candidate peptide was not present in the matched normal tissue DNA of the patient, it was considered a mutant peptide with somatic mutation properties, and a mutant peptide library was constructed for subsequent analysis. S4. MHC affinity prediction: Bioinformatics tools were used to predict the binding affinity of mutant peptides to the MHC molecules of the patient's HLA type, and high-affinity mutant peptides were screened as candidate neoantigens and synthesized in vitro. S5. Immunogenicity verification: The synthesized candidate neoantigen peptides were loaded onto antigen-presenting cells and co-cultured with peripheral blood T cells. The activation and cytokine secretion levels of T cells were detected by ELISPOT and flow cytometry to verify the immunogenicity of the candidate neoantigens and screen for tumor neoantigens with immunogenicity.

[0010] Furthermore, the tumor tissue sample in S1 is a solid colorectal cancer tumor tissue, and the purity of the gp96-tumor polypeptide complex is identified by SDS-PAGE, silver staining, reversed-phase HPLC, and Western blot analysis of gp96 / GRP94 monoclonal antibody.

[0011] Furthermore, the concentration of trifluoroacetic acid in S2 is 0.1%, and the peptide is purified using a C18 Stage-Tip desalting column, and the peptide concentration is measured to provide a standardized sample for subsequent mass spectrometry analysis.

[0012] Furthermore, the LC-MS / MS analytical conditions in S3 were as follows: 0.1% formic acid (FA) was used as mobile phase A, and 80% acetonitrile (ACN) containing 0.1% formic acid was used as mobile phase B for gradient elution. The purified peptide was dissolved in mobile phase A containing 0.1% formic acid, pre-column desalted and enriched online, and then entered the analytical column. The flow rate of the analytical column was kept constant at 600 nL / min, the gradient program was 0-66 minutes, and the concentration of mobile phase B increased linearly from 4% to 95%. The mass spectrometer adopted data-dependent acquisition mode, with a resolution of 70,000 m / z for the first-stage mass spectrometry and a scan range of 100-1500 m / z; and a resolution of 17,500 m / z for the second-stage mass spectrometry and a collision energy of 28 m / z.

[0013] Furthermore, the parameters analyzed by the PEAKS Studio software in S3 are as follows: fixed modification is cysteine ​​alkylation, variable modification includes methionine oxidation and N-terminal acetylation of peptide, non-specific enzyme digestion method, mass tolerance of 20 ppm and 0.02 Da for primary and secondary mass spectrometry, respectively, and the alignment database is the human reference proteome database; the identified peptide sequence is compared with the dbSNP public single nucleotide polymorphism database to exclude peptides that may originate from known germline polymorphisms and obtain candidate peptides; peripheral blood DNA matched by the patient is extracted for sequencing verification, and the gene sequence corresponding to the candidate peptide is compared with the sequencing results. If the sequence does not exist in the peripheral blood DNA of normal tissue matched by the patient, it is considered to be a mutant peptide from somatic mutation and is included in the mutant peptide library for subsequent MHC affinity prediction.

[0014] Furthermore, the bioinformatics tool in S4 is NetMHCpanEL 4.1 of the IEDB platform, with a median binding percentile value ≤ 0.5 as the criterion for determining strong binding candidates of MHC class I molecules.

[0015] Furthermore, in S5, the antigen-presenting cells are dendritic cells, and the criteria for determining immunogenicity are: the number of IFN-γ spots / positive cells in the candidate peptide stimulation well must be significantly higher than the baseline level of the blank control group and the level of the negative control group, and at least twice that of the negative control group.

[0016] Secondly, the present invention provides an application based on gp96 tumor neoantigen, wherein the immunogenic tumor neoantigen screened by the method can be used to prepare personalized cancer vaccines or as target molecules for adoptive T-cell therapy.

[0017] Furthermore, the screened neotumor antigens include one or more of HAVCR2_F39V, MUC3A_I29T, TMEM185B_A42G, and PLK1_A520T.

[0018] The present invention has the following beneficial effects: 1. In this invention, the prediction accuracy is significantly improved. Using gp96 as a natural bioaccumulator, the captured peptides include products in the natural antigen processing pathway of tumor cells. The enriched peptides have higher potential immunological significance. Combined with MHC affinity prediction and immunogenicity function verification, the true positive rate of neoantigens is greatly improved. In the experiment on colon cancer samples, 4 out of the 7 predicted candidate neoantigens can significantly activate the secretion of IFN-γ by the patient's autologous T cells, and the positive rate of immunogenicity verification is high.

[0019] 2. This invention enables personalized target screening, combining the patient's own tumor tissue samples, HLA typing, and autologous peripheral blood T cells throughout the process. The screened neoantigens are highly matched with the patient's immune background, making them suitable for personalized cancer vaccines and adoptive T-cell therapy. Compared to the cumbersome operation of traditional MHC enrichment methods that require the use of specific antibodies against specific HLA types, this invention utilizes the high conservation of gp96, eliminating the need for antibody enrichment against specific HLA types. It can extract tumor antigen peptides covering multiple HLA types from tumor tissue in a single step. The experimental method is simple and not limited by MHC typing, significantly expanding the clinically applicable population.

[0020] 3. In this invention, the experimental process is efficient and cost-controllable. The purification process of the gp96-peptide complex is simple and highly specific, and can directly enrich naturally processed antigen peptides from tumor tissues, which greatly reduces the sample size required for subsequent functional verification. Combined with high-throughput mass spectrometry, peptide sequences can be quickly identified. Compared with traditional sequencing methods that require large-scale synthesis of peptides for verification, the workload of this invention is significantly reduced, providing a feasible technical path for the rapid implementation of personalized immunotherapy. Attached Figure Description

[0021] Figure 1 Immunogenicity assessment of neoantigen peptides (ELISPOT assay spot pattern).

[0022] Figure 2 This is a bar chart showing the secretion of IFN-γ by T cells stimulated by peptides. Detailed Implementation

[0023] The technical solutions of the present invention will be clearly and completely described below with reference to the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0024] It should be noted that all raw materials used in the following experiments are commercially available.

[0025] After mechanical homogenization and thorough mixing, the same colon cancer tumor tissue block was divided into three portions, which were used for Comparative Example 1, Comparative Example 2, and this embodiment, respectively.

[0026] Example 1: Isolation and purification of gp96-tumor polypeptide complex: Tumor tissue samples from colorectal cancer patients were homogenized and centrifuged. The supernatant was used as the total protein extract. The total protein extract was loaded onto a ConA-Sepharose (GE) affinity chromatography column and the bound protein was eluted with 8% α-methyl glucoside. The glycoside eluent was loaded onto a HiTrapQ (GE chromatography system) anion exchange chromatography column for secondary purification. The purified product was collected, concentrated, and sterilized to obtain the gp96-tumor polypeptide complex.

[0027] The gp96 protein was identified by Western blot using gp96 / GRP94 monoclonal antibody (Santa Cruz). The purity of the complex was determined by SDS-PAGE, silver staining, and reversed-phase HPLC to ensure that the purity of gp96 protein was ≥90%.

[0028] Example 2: Dissociation and purification of gp96-bound peptides: The initial concentration of the gp96-tumor peptide complex in Example 1 was determined. After reductive alkylation of the sample, trifluoroacetic acid was added to a concentration of 0.1%, and the mixture was incubated for a period of time to dissociate the peptides from gp96. The dissociated peptide mixture was loaded onto a C18 Stage-Tip desalting column for desalting and purification of low molecular weight peptides. After purification, the peptide concentration was normalized to 1 μg / μL for subsequent mass spectrometry analysis.

[0029] Example 3: LC-MS / MS identification and sequence analysis of peptides: The standardized peptide sample from Example 2 was dissolved in mobile phase A containing 0.1% formic acid. After online desalting and enrichment on a pre-column (150 μm × 50 mm, Reprosil-Pur120C18-AQ, 3 μm), it was separated into an analytical column (150 μm × 170 mm, Reprosil-Pur120C18-AQ, 1.9 μm). The flow rate was set to 600 nL / min, and gradient elution was performed using 0.1% formic acid aqueous solution (phase A) and 80% acetonitrile containing 0.1% formic acid (mobile phase B). The elution gradient was 0–66 min, and the mobile phase B increased from 4% to 95%.

[0030] The separated peptides were sequentially sampled by a mass spectrometer. A data-dependent acquisition mode was used, with a first-stage mass spectrometry resolution of 70,000 m / z and a scan range of 100–1500 m / z. The top 20 ions with the highest intensities were selected for fragmentation. The second-stage mass spectrometry resolution was 17,500 m / z and the collision energy was 28 m / z, which yielded accurate mass and fragment spectra of the peptides.

[0031] Mass spectrometry data were analyzed using PEAKS Studio software. During the search, the fixed modification was set to cysteine ​​alkylation, and the variable modifications included methionine oxidation and N-terminal acetylation of the peptide. The enzyme digestion method was non-specific. The mass tolerances of the primary and secondary mass spectrometers were 20 ppm and 0.02 Da, respectively. The mass spectra were compared with the human reference proteome database (proteome_Homo_sapiens(9606)_reviewed) to identify the amino acid sequence of the peptide.

[0032] Example 4: Screening of mutant peptides and MHC affinity prediction: The peptide sequences identified in Example 3 were compared with the dbSNP public single nucleotide polymorphism database to exclude peptides that may originate from known germline polymorphisms, thus obtaining candidate peptides. Peripheral blood DNA matched by the patient was extracted and sequenced for verification. The gene sequences corresponding to the candidate peptides were compared with the sequencing results. If the sequence was not present in the peripheral blood DNA, it was considered a mutant peptide originating from somatic mutations and included in the mutant peptide library for subsequent MHC affinity prediction. The NetMHCpanEL 4.1 tool on the IEDB platform was used, with the mutant peptide sequence and the patient's HLA class I typing results as input parameters, to predict the binding affinity of the mutant peptide to MHC class I molecules. A median binding percentile value ≤ 0.5 was used as the strong binding criterion. Seven high-affinity candidate neoantigen peptides were predicted, and their details are shown in Table 1.

[0033] Table 1. Prediction results of candidate neoantigens for colorectal cancer

[0034] Example 5: Immunogenicity Validation of Candidate Neoantigen: The immunogenicity of the candidate neoantigen was validated using the IFN-γ ELISPOT method based on the patient's autologous PBMCs. Peripheral blood was collected from the colon cancer patient, and autologous PBMCs were isolated. ELISPOT plates pre-coated with anti-human IFN-γ antibody were blocked with culture medium containing 10% fetal bovine serum, with 3 × 10⁶ cells / well inoculated per well. 5 Each autologous PBMC suspension was used; the experimental group was supplemented with candidate neoantigen synthetic peptides at a concentration of 10 μg / mL, and a corresponding wild-type non-mutant peptide was set up as a control for each neoantigen. The positive control was supplemented with phytohemagglutinin, and the negative control was untreated cell culture.

[0035] The culture plates were incubated at 37°C and 5% CO2 for 48 hours. The formation of specific spots was detected and statistically analyzed. The criteria for judgment were: the spot count of the neoantigen peptide-stimulated well was significantly higher than that of the blank control and significantly higher than that of the corresponding wild-type non-mutant peptide well.

[0036] Comparative Example 1: Tumor tissue and adjacent normal tissue from the same colorectal cancer patient were subjected to WES genome sequencing. MHC affinity was predicted using NetMHCpan, and 52 candidate neoantigens were screened out. Immunogenicity was verified using the same ELISPOT method as in Example 5. Only 6 were positive, with a positive rate of 11.5%.

[0037] Comparative Example 2: Tumor tissue from the same colorectal cancer patient was used to enrich MHC class I molecules by immunoprecipitation, and binding peptides were obtained by acid elution. After identification by LC-MS / MS, NetMHCpanEL 4.1 predicted 11 candidate neoantigens, and ELISPOT verification showed that only 5 were positive, with a positive rate of 45.5%.

[0038] Performance testing: The gp96-based tumor neoantigen prediction method and its application prepared in Examples 1, 2, 3, 4, 5, Comparative Example 1 and Comparative Example 2 were tested.

[0039] Performance testing: The performance of the gp96-based tumor neoantigen prediction method and its application samples provided in Examples 1-5 and Comparative Examples 1-2 were tested respectively. The test data obtained from the examples are as follows: Figure 1 , Figure 2 As shown: Experimental results are as follows Figure 1 , Figure 2 As shown: Figure 1 The images show representative blot patterns obtained using ELISPOT. In each group of four wells, the first two wells were stimulated by the mutant peptide, and the last two wells were stimulated by the corresponding wild-type peptide. The results showed that the number of IFN-γ blots formed by stimulation with the mutant peptides HAVCR2_F39V, MUC3A_I29T, TMEM185B_A42G, and PLK1_A520T was significantly higher than that of their corresponding blank controls and wild-type non-mutant peptides, suggesting that these four candidate neoantigens possess mutation-specific immunogenicity. However, the number of blots stimulated by the remaining mutant peptides (C5_Q1448E, PRKCSH_Q435E, PRKCSH_T444N) did not differ significantly from the corresponding wild-type peptides, indicating that their immunogenicity was weak or non-specific.

[0040] Figure 2 The quantitative statistical results of the number of IFN-γ positive spots are shown. The horizontal axis represents each candidate neoantigen and its corresponding wild-type peptide, and the vertical axis represents the number of spots per 3 × 10⁻⁶ positive spots. 5The number of IFN-γ positive spots in each PBMC. As shown in the figure, the number of IFN-γ spots in the wells stimulated by the four mutant peptides HAVCR2_F39V, MUC3A_I29T, TMEM185B_A42G, and PLK1_A520T was higher than that in their corresponding wild-type controls, and more than twice that of the wild-type peptides. The numbers of the other three mutant peptides were similar to those of their corresponding wild-type peptides, with no significant difference.

[0041] In summary, this invention significantly outperforms the comparative methods in terms of the accuracy of tumor neoantigen prediction, the ability to screen personalized targets, and the efficiency and cost control of the experimental process, fully verifying the advanced nature and feasibility of the proposed technical solution.

[0042] In the description of this specification, references to terms such as "an embodiment," "example," "specific example," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0043] The preferred embodiments of the present invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to the specific implementations described. Clearly, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention. The invention is limited only by the claims and their full scope and equivalents.

Claims

1. A method for predicting tumor neoantigens based on gp96, characterized in that, The gp96-based tumor neoantigen prediction method includes the following steps: Isolation and purification of S1.gp96-tumor polypeptide complex: Total protein was extracted from patient tumor tissue samples, purified by two steps of ConA affinity chromatography and HiTrap Q ion exchange, concentrated and sterilized, and gp96-tumor polypeptide complex was specifically isolated. S2. Dissociation and purification of peptides: The gp96-tumor peptide complex was incubated under acidic conditions containing trifluoroacetic acid to dissociate the peptides from gp96. The dissociated peptide mixture was desalted and purified into low molecular weight peptides by C18 solid phase extraction column. S3. Mass Spectrometry Identification and Sequence Analysis of Mutant Peptides: The purified peptides were analyzed by liquid chromatography-tandem mass spectrometry (LC-MS / MS). The mass spectrometry data were analyzed using PEAKS Studio software to identify the amino acid sequence of the peptides. The identified peptide sequences were compared with the dbSNP public single nucleotide polymorphism database to exclude peptides with known germline polymorphisms and obtain candidate peptides. Peripheral blood DNA from patients was extracted for sequencing verification. If the gene sequence corresponding to the candidate peptide was not found in the peripheral blood DNA of the patient's matched normal tissue, the peptide was considered to be a mutant peptide from somatic cell mutation. Such peptides were included in the mutant peptide library for subsequent neoantigen prediction analysis. S4. MHC affinity prediction: Bioinformatics tools were used to predict the binding affinity of mutant peptides to the MHC molecules of the patient's HLA type, and high-affinity mutant peptides were screened as candidate neoantigens and synthesized in vitro. S5. Immunogenicity verification: The synthesized candidate neoantigen peptides were loaded onto antigen-presenting cells and co-cultured with peripheral blood T cells. The activation and cytokine secretion levels of T cells were detected by ELISPOT and flow cytometry to verify the immunogenicity of the candidate neoantigens and screen for tumor neoantigens with immunogenicity.

2. The method for predicting tumor neoantigens based on gp96 according to claim 1, characterized in that, The tumor tissue sample in S1 is a solid colorectal cancer tumor tissue. The purity of the gp96-tumor polypeptide complex was identified by SDS-PAGE, silver staining, reversed-phase HPLC, and Western blot analysis of gp96 / GRP94 monoclonal antibody.

3. The method for predicting tumor neoantigens based on gp96 according to claim 1, characterized in that, The concentration of trifluoroacetic acid in S2 was 0.1%. The peptide was purified using a C18 Stage-Tip desalting column, and the peptide concentration was measured to provide a standardized sample for subsequent mass spectrometry analysis.

4. The method for predicting tumor neoantigens based on gp96 according to claim 1, characterized in that, The analytical conditions for LC-MS / MS in S3 were as follows: 0.1% formic acid (FA) was used as mobile phase A, and 80% acetonitrile (ACN) containing 0.1% formic acid was used as mobile phase B. Gradient elution was performed. The purified peptide was dissolved in mobile phase A containing 0.1% formic acid, pre-column desalted and enriched online, and then entered the analytical column. The flow rate of the analytical column was kept constant at 600 nL / min, and the gradient program was 0-66 minutes. The concentration of mobile phase B increased linearly from 4% to 95%. The mass spectrometer adopted data-dependent acquisition mode. The resolution of the first-stage mass spectrometer was 70,000, and the scan range was 100-1500 m / z. The resolution of the second-stage mass spectrometer was 17,500, and the collision energy was set to 28.

5. The method for predicting tumor neoantigens based on gp96 according to claim 1, characterized in that, The parameters analyzed by the PEAKSStudio software in S3 are as follows: fixed modification is cysteine ​​alkylation, variable modifications include methionine oxidation and N-terminal acetylation of peptides, enzyme digestion method is non-specific, and the mass tolerances of primary and secondary mass spectrometry are 20 ppm and 0.02 Da, respectively. The identified peptide sequences are compared with the dbSNP public single nucleotide polymorphism database to obtain candidate peptides. Peripheral blood DNA matched by the patient is extracted and sequenced for verification. The gene sequence corresponding to the candidate peptide is compared with the sequencing results. If the sequence does not exist in the normal tissue DNA matched by the patient, the peptide is considered to be a mutant peptide from somatic cell mutation. Such peptides are included in the mutant peptide library for subsequent neoantigen prediction analysis.

6. The method for predicting tumor neoantigens based on gp96 according to claim 1, characterized in that, In S4, the bioinformatics tool is NetMHCpanEL 4.1 on the IEDB platform, and the median binding percentile value ≤ 0.5 is used as the criterion for determining strong binding candidates of MHCI class molecules.

7. The method for predicting tumor neoantigens based on gp96 according to claim 1, characterized in that, In S5, the antigen-presenting cells are dendritic cells. The criteria for determining immunogenicity are: the number of IFN-γ spots / positive cells in the candidate peptide stimulation well must be higher than the baseline level of the blank control group without peptide stimulation and the level of the wild-type non-mutant peptide negative control group, and at least twice that of the negative control group.

8. An application based on gp96 tumor neoantigen according to any one of claims 1-7, characterized in that, The immunogenic neoantigens screened by this method can be used to prepare personalized cancer vaccines or as target molecules for adoptive T-cell therapy.

9. The application based on gp96 tumor neoantigen according to claim 8, characterized in that, The selected neotumor antigens include one or more of HAVCR2_F39V, MUC3A_I29T, TMEM185B_A42G, and PLK1_A520T.

Images