PROTEOGENOMICS-BASED METHOD FOR IDENTIFYING TUMOR-SPECIFIC ANTIGENS

Abstract

T cells, particularly CD8+ T cells, are known to be essential players in tumor eradication, as the presence of tumor-infiltrating lymphocytes (TILs) in various cancers is positively correlated with a good prognosis. To eliminate tumor cells, CD8+ T cells recognize tumor antigens, which are MHC class I-associated peptides present on the surface of tumor cells and expressed at very low or no levels on normal cells.This description describes a proteogenomic approach using RNA sequencing data from equivalent cancerous and normal mTEChi samples to identify non-tolerogenic tumor-specific antigens derived from (i) coding and non-coding regions of the genome, (ii) non-synonymous single-base mutations or short insertions / deletions and more complex rearrangements, as well as (iii) endogenous retroelements, which function independently of mutational burden or sample complexity.

Description

METHOD BASED ON PROTEOGENOMICS TO IDENTIFY TUMOR SPECIFIC ANTIGENS CROSS REFERENCE TO RELATED REQUESTS This application claims the benefits of United States Provisional Patent Application No. 62 / 724,760, filed August 30, 2018, which is incorporated herein by reference in its entirety. TECHNICAL FIELD The present invention relates generally to cancer and, more specifically, to the identification of tumor antigens useful for T cell-based cancer immunotherapy. BACKGROUND OF THE ART CD8 T cells are known to be essential players in tumor eradication, as the presence of tumor-infiltrating lymphocytes (TILs) in various cancers positively correlates with good prognosis and response to immune checkpoint inhibitors.1 ,2. To kill tumor cells, CD8 T cells recognize tumor antigens, which are abnormal MHC I-associated peptides (MAPs) presented by tumor cells. Since CD8 T cells recognize MHC I-associated peptides (MAPs), the most important unanswered question is the nature of the MAPs recognized by CD83 TILs. Knowing that the abundance of CD8 TILs correlates with the mutation load of tumors, the dominant paradigm holds that CD8 TILs recognize mutated tumor-specific antigens (mTSAs), commonly called neoantigens2,4,5. The superior immunogenicity of mTSAs is attributed to their selective expression in tumors that minimizes the risk of immune tolerance6. However, some TILs have been shown to recognize cancer-restricted wild-type MAPs7 which will be referred to as aberrantly expressed TSAs (aeTSAs). aeTSAs can derive from a variety of cis- or trans-acting genetic and epigenetic changes that lead to the transcription and translation of genomic sequences that are not expressed in normal cells, such as endogenous retroelements (EREs)8-10. Considerable effort is being devoted to discovering processable TSAs that can be used in therapeutic cancer vaccines. The most common strategy relies on reverse immunology: i) exorn sequencing is performed on tumor cells to identify mutations, and ii) MHC binding prediction software tools are used to identify which mutated MAPs might be good MHC binders11,12. Although reverse immunology can enrich TSA candidates, at least 90% of these candidates are false positives5,13 because available computational methods can predict MHC binding, but cannot predict other steps involved in MAP processing14, fifteen. To overcome this limitation, some studies include mass spectrometry (MS) analyzes in their TSA16 discovery pipeline, thereby providing a rigorous molecular definition of various TSAs17,18. However, the performance of these approaches has been extremely poor: in melanoma, one of the most mutated tumor types, MS has validated an average of 2 TSAs per individual tumor19, while only a handful of TSAs have been found for other types of cancer15. The paucity of TSA is puzzling because injection of TILs or immune checkpoint inhibitors would not cause tumor regression if tumors did not express immunogenic antigens20. It was assumed that exonic mutation-based approaches had failed to identify TSAs because they did not take into account two crucial elements. First, these approaches focus solely on mTSAs and neglect aeTSAs, essentially because there is currently no method for high-throughput identification of aeTSAs. This represents a major shortcoming because, although mTSAs are private antigens, aeTSAs would be preferred targets for vaccine development, since they can be shared by multiple tumors.7,9. Second, focusing on exorna as the only source of UT is too restrictive. The exorna (ie all protein-coding genes) represents only 2% of the human genome, while up to 75% of the genome can be transcribed and potentially translated22. Thus, there is a need for novel approaches to identify tumor antigens that can be used for T cell-based cancer immunotherapy. Acute lymphoblastic leukemia (ALL) is a malignant transformation and proliferation of lymphoid progenitor cells in bone marrow, blood, and extramedullary sites. Although 80% of ALL occurs in children, it is a devastating disease when it occurs in adults. Within the United States, the incidence of ALL is estimated to be 1.6 per 100,000 population. Although dose escalation strategies have led to significant improvement in pediatric patient outcomes, the prognosis for the elderly remains very poor. Despite a high response rate to induction chemotherapy, only 30-40% of adult ALL patients will achieve long-term remission. Therefore, there is a need for novel approaches to the treatment of ALL. Lung cancer, a highly invasive, rapidly metastatic and prevalent cancer, is the leading cancer cause of death in men and women in the United States of America (USA). Approximately 90% of lung cancer cases are caused by smoking and the use of tobacco products. However, other factors such as radon gas, asbestos, exposure to air pollution, and chronic infections can contribute to lung carcinogenesis. In addition, multiple inherited and acquired mechanisms of susceptibility to lung cancer have been proposed. Lung cancer is divided into two broad histologic classes, which grow and spread differently: small cell lung carcinomas (SCLC) and non-small cell lung carcinomas (NSCLC). Treatment options for lung cancer include surgery, radiation therapy, chemotherapy, and targeted therapy. Despite improvements in diagnosis and therapy made over the last 25 years, the prognosis for patients with lung cancer remains poor. Responses to current standard therapies are poor except for the most localized cancers. Therefore, there is a need for novel approaches to the treatment of lung cancer. The present description refers to a series of documents, the contents of which are incorporated by reference in their entirety in the present description. BRIEF DESCRIPTION OF THE INVENTION This description provides the following elements from 1 to 75: 1. A method for identifying a candidate tumor antigen in a sample of tumor cells, the method comprising: (a) generating a database of tumor-specific proteomes by: (i) extracting a set of subsequences (k-mers) comprising at least 33 base pairs of tumor RNA sequences; (ii) comparing the set of tumor subsequences from (i) with a set of corresponding control subsequences comprising at least 33 base pairs extracted from normal cell RNA sequences; (iii) extracting the tumor subsequences that are absent in the corresponding control subsequences, to thereby obtain tumor-specific subsequences; and (iv) translating the tumor-specific subsequences in silico, to thereby obtain the tumor-specific proteome database; (b) generating a customized tumor proteome database by: (i) comparing the tumor RNA sequences to a reference genome sequence to identify single base mutations in said tumor RNA sequences; (ii) inserting the single base mutations identified in (i) into the reference genome sequence, thereby creating a personalized tumor genome sequence; (iii) translating in silico the transcripts encoding expressed proteins from said personalized tumor genome sequence, to thereby obtain the personalized tumor proteome database; (c) comparing the sequences of the major histocompatibility complex (MHC)-associated peptides (MAP) of said tumor with the sequences of the tumor-specific proteome database of (a) and the customized tumor proteome database of (b) to identify MAPs; and (d) identifying a candidate tumor antigen among the MAPs identified in (c), wherein a candidate tumor antigen is a peptide whose sequence and / or coding sequence is overexpressed or overrepresented in tumor cells relative to normal cells. 2. The method of item 1, wherein the aforementioned method further comprises (1) isolating and sequencing the major histocompatibility complex (MHC)-associated peptides (MAP) from the tumor cell sample, and / or (2) performing the sequencing of the complete transcriptome in the sample of tumor cells, to obtain the RNA sequences of the tumor. 3. The method of item 2, wherein said isolating MAPs comprise (i) releasing said MAPs from said cell sample by mild acid treatment; and (ii) subjecting the released MAPs to chromatography. 4. The method of item 3, wherein said method further comprises filtering the released peptides with a size exclusion column prior to said chromatography. 5. The method of any of items 1 to 4, wherein said subsequences comprise 33 to 54 base pairs. 6. The method of any of items 1 to 5, further comprising assembling overlapping tumor-specific subsequences into longer tumor subsequences (contigs). 7. The method of item 6, wherein said size exclusion column is capped at approximately 3000 Da. 8. The method of any of items 1 to 7, wherein said MAP sequencing comprises subjecting the isolated MAPs to mass spectrometry (MS) sequencing analysis. 9. The method of any of items 1 to 8, wherein said method further comprises generating a custom database of normal proteomes by using the corresponding normal cells. 10. The method of item 9, wherein said identification in (d) comprises excluding said MAP if its sequence is detected in the normal custom proteome database. 11. The method of any of items 1 to 10, wherein the method further comprises generating 24 or 39 nucleotide k-mer databases from said tumor RNA sequences and normal cell RNA sequences to obtain a tumor k-mer database and a normal k-mer database; and compare' the tumor k-mer database and a normal k-mer database with k-mer of 24 or 39 nucleotides derived from the sequence that encodes MAP, where an overexpression or overrepresentation of the k-mer derived from the MAP-encoding sequence in said tumor k-mer database relative to said normal k-mer database is indicative that the corresponding MAP is a candidate tumor antigen. 12. The method of item 11, wherein the k-mer derived from the MAP coding sequence is at least 10-fold overexpressed or overrepresented in said tumor k-mer database relative to said k-mer database normal. 13. The method of item 11 or 12, wherein the k-mer derived from the MAP coding sequence is absent from said normal k-mer database. 14. The method of any of items 1 to 13, wherein said method comprises: (a) isolating and sequencing MAP in a sample of tumor cells; (b) performing whole transcriptome sequencing on said sample of tumor cells, to thereby obtain tumor RNA sequences; (c) generating a database of tumor-specific proteomes by: (i) extracting a set of subsequences comprising at least 33 nucleotides from said tumor RNA sequences; (ii) comparing the set of tumor subsequences from (i) with a set of corresponding control subsequences comprising at least 33 nucleotides extracted from normal cell RNA sequences; (iii) extracting the immoral subsequences that are absent, or underexpressed by at least 4-fold, in the corresponding control subsequences, to thereby obtain tumor-specific subsequences; and (iv) translating the tumor-specific subsequences in silico, to thereby obtain the tumor-specific proteome database; (d) generating a customized tumor proteome database by: (i) comparing the tumor RNA sequences to a reference genome sequence to identify single base mutations in said tumor RNA sequences; (ii) inserting the single base mutations identified in (i) into the reference genome sequence, thereby creating a personalized tumor genome sequence; (iii) translating in silico the transcripts encoding expressed proteins from said personalized tumor genome sequence, to thereby obtain the personalized tumor proteome database; (e) generating a customized normal proteome database by: (i) comparing RNA sequences from normal cells to a reference genome sequence to identify single base mutations in said normal RNA sequences; (ii) inserting the single base mutations identified in (i) into the reference genome sequence, thereby creating a custom normal genome sequence; (iii) translating in silico the expressed protein-encoding transcripts from said custom normal genome sequence, to thereby obtain the custom normal proteome database; (f) generating a normal and tumor k-mer database (i) extracting a set of subsequences comprising at least 24 nucleotides from said normal cell RNA sequences and said Immoral RNA sequences; (g) comparing the sequences of the MAPs obtained in (a) with the sequences of the tumor-specific proteome database of (c) and the customized tumor proteome database of (d) to identify the MAPs; and (h) identifying a candidate tumor antigen among the MAPs identified in (f), wherein a candidate tumor antigen corresponds to a MAP (1) whose sequence is not present in the custom normal proteome database; and (2) (i) whose sequence is present in the personalized tumor proteome database; and / or (ii) whose coding sequence cccznn / i znz / B / v is overexpressed or overrepresented in said tumor k-mer database relative to said normal k-mer database. 15. The method of any of items 1 to 14, wherein said method further comprises selecting MAPs that are 8 to 11 amino acids in length. 16. The method of any of items 1 to 15, wherein said normal cells are thymic cells. 17. The method of item 16, wherein said thymic cells are medullary thymic epithelial cells (mTEC). 18. The method of any of items 1 to 17, further comprising comparing the coding sequence of said candidate tumor antigen with sequences from normal tissues. 19. The method of any of items 1 to 18, wherein said MAPs are 8 to 11 amino acids in length. 20. The method of any of items 1 to 19, further comprising assessing the binding of the candidate tumor antigen to an MHC molecule. 21. The method of item 20, wherein said binding is evaluated using an MHC binding prediction algorithm. 22. The method of any of items 1 to 21, further comprising evaluating the frequency of T cells that recognize the candidate tumor antigen in a cell population. 23. The method of item 22, wherein the frequency of T lymphocytes recognizing the candidate tumor antigen is assessed by using multimeric MHC class I molecules comprising said candidate tumor antigen in its peptide-binding groove. 24. The method of any of items 1 to 23, further comprising evaluating the ability of the candidate tumor antigen to induce T cell activation. 25. The method of item 24, wherein the ability of the candidate tumor antigen to induce T cell activation is assessed by measuring cytokine production by T cells in contact with cells having said candidate tumor antigen bound to MHC molecules class I on its cell surface. 26. The method of item 25, wherein said cytokine production comprises interferon-gamma (ΙΕΝ-γ) production. 27. The method of any of items 1 to 26, further comprising evaluating the ability of said candidate tumor antigen to induce T-cell mediated immoral cell death and / or to inhibit tumor growth. 28. A tumor antigen peptide identified by the method defined in any of items 1 to 27. 29. A tumor antigen peptide comprising or consisting of one of the amino acid sequences set forth in any of SEQ ID NO: 1-39. 30. The tumor antigen peptide of item 29, comprising or consisting of one of the amino acid sequences set forth in any of SEQ ID NO: 17-39. 31. The tumor antigen peptide of item 30, wherein said tumor antigen peptide is a leukemia tumor antigen peptide and comprises or consists of one of the amino acid sequences set forth in any of SEQ ID NO: 17-28. 32. The tumor antigen peptide of item 31, wherein said leukemia is B-cell acute lymphoblastic leukemia (B-ALL). 33. The tumor antigen peptide of item 31 or 32, wherein said tumor antigen peptide binds to a human leukocyte antigen (HLA) HLA-A*02:01 allele and comprises or consists of one of the amino acid sequences set forth in any of SEQ ID NO: 17-19, 27 and 28. 34. The tumor antigen peptide of point 31 or 32, wherein said tumor antigen peptide binds to a human leukocyte antigen (HLA) of the HLA-B*40:01 allele and comprises or consists of the amino acid sequence set forth in SEQ ID NO: 20. 35. The tumor antigen peptide of item 31 or 32, wherein said tumor antigen peptide binds to a human leukocyte antigen (HLA) of the HLA-A* 11:01 allele and comprises or consists of one of the amino acid sequences set forth in any of SEQ ID NO: 21-23. 36. The tumor antigen peptide of point 31 or 32, wherein said tumor antigen peptide binds to a human leukocyte antigen (HLA) of the HLA-B*08:01 allele and comprises or consists of the amino acid sequences set forth in SEQ ID NO: 24 or 25. 37. The tumor antigen peptide of point 31 or 32, wherein said tumor antigen peptide binds to a human leukocyte antigen (HLA) of the HLA-B*07:02 allele and comprises or consists of the amino acid sequence set forth in SEQ ID NOT: 26. 38. The tumor antigen peptide of item 30, wherein said tumor antigen peptide is a lung tumor antigen peptide and comprises or consists of one of the amino acid sequences set forth in any of SEQ ID NO: 29-39. 39. The tumor antigen peptide of item 38, wherein said lung tumor is non-small cell lung cancer (NSCLC). 40. The tumor antigen peptide of item 38 or 39, wherein said tumor antigen peptide binds to a human leukocyte antigen (HLA) of the HLA-A* 11:01 allele and comprises or consists of one of the amino acid sequences set forth in any of SEQ ID NO: 29-35. 41. The tumor antigen peptide of point 38 or 39, wherein said tumor antigen peptide binds to a human leukocyte antigen (HLA) of the HLA-B*07:02 allele and comprises or consists of the amino acid sequence set forth in SEQ ID NO: 36. 42. The tumor antigen peptide of point 38 or 39, wherein said tumor antigen peptide binds to a human leukocyte antigen (HLA) of the HLA-A*24:02 allele and comprises or consists of the amino acid sequences set forth in SEQ ID NO: 38 or 39. 43. The tumor antigen peptide of point 38 or 39, wherein said tumor antigen peptide binds to a human leukocyte antigen (HLA) of the HLA-C*07:01 allele and comprises or consists of the amino acid sequence set forth in SEQ ID NO: 37. 44. The tumor antigen from any of items 29-43, which is derived from a non-protein-coding region of the genome. 45. The tumor antigen of point 44, wherein said non-protein-coding region of the genome is an intergenic region, an intronic region, a 5' untranslated region (5' UTR), a 3' untranslated region (3' UTR ) or an endogenous retroelement (ERE). 46. ​​A nucleic acid encoding the tumor antigen peptide of any of items 28-45. 47. The nucleic acid of item 46, which is an mRNA or a viral vector. 48. A liposome comprising the tumor antigen peptide of either item 2845 or the nucleic acid of item 46 or 47. 49. A composition comprising the tumor antigen peptide of any of items 28-45, the nucleic acid of item 46 or 47, or the liposome of item 48, and a pharmaceutically acceptable carrier. cccznn / i znz / B / v 50. A vaccine comprising the immoral antigenic peptide of either item 2845, the nucleic acid of item 46 or 47, the liposome of item 48, or the composition of item 49, and an adjuvant. 51. An isolated major histocompatibility complex (MHC) class I molecule comprising the tumor antigen peptide of any of elements 28-45 in its peptide-binding groove. 52. The isolated MHC class I molecule of item 51, which is in the form of a multimer. 53. The MHC class I molecule isolated from item 52, wherein said multimer is a tetramer. 54. An isolated cell comprising the tumor antigen peptide of any of items 28-45. 55. An isolated cell expressing on its surface major histocompatibility complex (MHC) class I molecules comprising the tumor antigen peptide from any of sites 28-45 in its peptide-binding groove. 56. The cell at point 55, which is an antigen presenting cell (APC). 57. The cell of item 56, wherein said APC is a dendritic cell. 58. A T cell receptor (TCR) that specifically recognizes MHC class I molecule isolated from any of items 51-53 and / or MHC class I molecules expressed on the cell surface from any of items 54-57. 59. An isolated CD8+ T cell expressing the TCR from point 58 on its cell surface. 60. A cell population comprising at least 0.5% CD8+ T lymphocytes as defined in item 59. 61. A method of treating cancer in a subject comprising administering to the subject an effective amount of: (i) the tumor antigen peptide of any of items 28-45; (ii) the nucleic acid of point 46 or 47; (iii) the liposome of item 48; (iv) the composition of point 49; (v) the 50 point vaccine; (vi) the cell of any of items 54-57; (vii) CD8+ T lymphocytes from point 59; or (viii) the cell population of point 60. 62. The method of item 61, wherein said cancer is leukemia. 63. The method of item 62, wherein said leukemia is acute lymphoblastic cell leukemia B (B-ALL). 64. The method of item 61, wherein said cancer is lung cancer. 65. The method of item 64, wherein said lung tumor is non-small cell lung cancer (NSCLC). 66. The method of any one of items 61-65, further comprising administering at least one additional antitumor agent or therapy to the subject. 67. The method of item 66, wherein said at least one additional antitumor agent or therapy is a chemotherapeutic agent, immunotherapy, an immune checkpoint inhibitor, radiotherapy, or surgery. 68. Use of: (i) the tumor antigen peptide of any of items 28-45; (ii) the nucleic acid of point 46 or 47; (iii) the liposome of item 48; (iv) the composition of point 49; (v) the 50 point vaccine; (vi) the cell of any of items 54-57; (vii) CD8+ T lymphocytes from point 59; or (viii) the cell population of site 60, to treat cancer in a subject. 69. Use of: (i) the tumor antigen peptide of any of items 28-45; (ii) the nucleic acid of point 46 or 47; (iii) the liposome of item 48; (iv) the composition of point 49; (v) the 50 point vaccine; (vi) the cell of any of items 54-57; (vii) CD8+ T lymphocytes from point 59; or (viii) the cell population of item 60, for the manufacture of a medicament for treating cancer in a subject. 70. The use of point 68 or 69, where said cancer is leukemia. 71. Use of item 70, wherein said leukemia is B-cell acute lymphoblastic leukemia (B-ALL). 72. The use of item 68 or 69, where said cancer is lung cancer. 73. Use of item 72, wherein said lung tumor is non-small cell lung cancer (NSCLC). 74. The use of any of items 68-73, further comprising the use of at least one additional antitumor agent or therapy. 75. The use of item 74, wherein said at least one additional antitumor agent or therapy is a chemotherapeutic agent, immunotherapy, an immune checkpoint inhibitor, radiotherapy, or surgery. Other objects, advantages and features of the present invention will become more apparent upon reading the following non-restrictive descriptions of specific embodiments thereof, given by way of example only with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS In the attached figures: Figures 1A-C show the targeted proteogenomic workflow for the identification of tumor-specific antigens (TSAs). Figures ΙΑ, B: Schematic detailing how the canonical cancer proteome (Fig. 1A) and cancer-specific proteome (Fig. IB) were constructed for each sample analyzed. Figure 1C: The combination of these two proteomes, termed the global cancer database, was used to identify MAP, and more specifically TSA, sequenced by liquid chromatography-MS / MS (LC-MS / MS) from two murine cell lines. well characterized samples, specifically CT26 and EL4, and seven human primary samples, specifically, four B-ALL and three lung tumor biopsies (n = 2-4 per sample). Statistics on each part of the global cancer database can be found in Tables 4a-b, while implementation details for constructing the cancer-specific proteome using k-mer profiling are presented in Figure 7. aa: amino acids, nts: nucleotides, th: sample-specific threshold at k-mer occurrence (see k-mer filtering and cancer-specific proteome generation section of Example 1 below). Figures 2A-D depict the results of experiments showing that the majority of TSAs are derived from translation of non-coding regions. Figure 2A: Flowcharts indicating the key validation steps involved in TSA discovery. Details about each step can be found in Figure 8. Figure 2B: Most TSA candidates derive from aberrantly expressed sequences. Bar graph showing the number of mTSA (m) and aeTSA (ae) candidates in the CT26 and EL4 tumor models. Figure 2C: Heatmap showing MCS expression for aeTSA candidates in 22 tissues / organs for which RNA-Seq data is publicly available (see Table 5). MCS expression for previously overexpressed TAA EL440,41 is shown as a control. Expression values ​​were normalized to rphm (reads per hundred million sequenced reads, see MCS Peripheral Expression section of Example 1 for details) and averaged across all available RNA-Seq experiments for each tissue. The bold squares indicate the tissues in which the relevant MCS was detected at rphm > 0. Tej. adip: adipose tissue, glán mam.: mammary gland and tej. adip. sub.: subcutaneous adipose tissue. Figure 2D: Most of the ae / mTSAs are derived from non-coding regions. Bar graphs depicting the number of TSAs derived from in-frame translation of coding exons cccznn / i znz / B / v (inside coding), out-of-frame translation of coding exons (outside coding), and translation regions supposedly non-coding (non-coding). Numbers within bars represent aeTSA / mTSA numbers. The percentages above the bars indicate the proportion of TSAs derived from atypical translation events, ie, TSAs that belong to the coding and non-coding categories. The characteristics of CT26 and EL4 TSA can be found in Tables la and b, respectively. Figures 3A-C are graphs showing that immunization against individual TSAs confers different degrees of protection against EL4 cells. Female C57BL / 6 mice were immunized twice with individual TSA-pulsed DCs as follows: (Figure 3A) two aeTSAs, (Figure 3B) two ERE TSAs (one aeTSA or one mTSA) and (Figure 3C) one mTSA. Mice were injected i.v. with 5 x 105^ live EL4 cells (black triangles) on day 0 and all surviving mice were boosted on day 150. Control groups were immunized with non-pulsed DC (black line). represents median survival. Statistical significance of the immunized vs. control groups was calculated by using a log-rank test, where ns means not significant (p > 0.05). 10 mice per group for peptide-specific immunization, 19 mice for control group. Figures 4A-D are graphs showing the frequency and secretion of IFN-γ by TSA-specific T cells in naïve and immunized mice. Figure 4A: Number of tetramer+CD8+ T cells per 106 CD8+ T cells in naïve mice. Each circle represents one mouse (n = 5 to 9 mice). The dotted line represents a frequency of 1 tetramer+ T cell per 106 CD8+ T cells. P values ​​were calculated using two-tailed Mann-Whitney tests (**; p < 0.01 and ***; p < 0.001). Figure 4B: Expansion of antigen-specific CD8+ T cells after immunization. Double enrichment for tetramers+CD8+ T cells was calculated by dividing the mean frequency in mice immunized with relevant (white bars) or irrelevant (gray bars) peptides by the mean frequency in naïve mice. Figures 4C, 4D: Sorted CD8+ T cells from immunized mice were incubated for 48 hours in the presence of peptide-pulsed irradiated splenocytes. Figure 4C: The frequency of IFN-γ-secreting antigen-specific cells is expressed as the mean frequency of spot-forming cells (CFS reported by 10 6 CD8 T cells on plate) in less immunized than in naïve mice. Three independent experiments with circles representing technical replicates. P-values ​​were calculated using unpaired two-tailed Student's t-tests, all p<0.05 (range: 0.0025 - 0.0143). Figure 4D: Functional avidity of antigen-specific T cells was calculated by normalizing the SFC frequency to the maximum value and calculating an EC50 for each peptide using a dose-response curve. Functional avidity values ​​for H7a and H13ase have been previously published and used for comparison purposes. Three independent experiments. In all relevant panels, the full horizontal lines and the numbers above each condition represent the mean values. Viral peptides used as controls are highlighted in grey. ^ Figures 5A-D are graphs showing that high expression of EL4-derived TSA is important but not sufficient to induce antileukemic responses. Figures 5A, B: Analysis of TSA expression at the RNA and peptide level was performed in EL4 cells injected at day 0 or 150, respectively. Figure 5A: Bar graph depicting the number of EL4 RNA-Seq reads that completely overlap with the MCS encoding each of the five EL4 TSAs. Figure 5B: TSA copy number per cell, estimated by PRM MS using 13 synthetic C-peptide analogs of the five EL4 TSAs. Three replicate EL4 cells by TSA. The average TSA copy number per cell is indicated on the left side of the graph. N.D.: not detected. Figure 5C: Expansion of TSA-specific TCD8+ cells after injection with live EL4 cells without prior immunization with peptide-pulsed DC. The double enrichment for tetramer+CD8+ T cells was calculated by dividing the mean frequency in EL4-injected mice by the mean frequency in naïve mice. Double enrichment for T cells recognizing viral peptides, which are not presented by EL4 cells, are shown as negative controls and are highlighted in grey. Figure 5D: C57BL / 6 female mice were immunized twice with 592 d irradiated EL4 cells (10,000 cGy) (blue line) or non-pulsed CD as control (black line) and then injected i.v. with 5 x 105 live EL4 cells, it represents median survival. 10 mice for immunization with irradiated EL4 cells, 19 mice for the control group. Figures 6A-C show that the majority of TSAs detected in human primary tumors are derived from translation of non-coding regions. Figure 6A: Most human TSAs are aeTSAs. Bar chart showing the number of aeTSA (ae) and mTSA (m) candidates in each primary sample tested. Figure 6B: Peripheral expression of human aeTSA candidates and TAA. Heatmap showing MCS expression for the 27 aeTSA and 24 Overexpressed TAAs, obtained from the Cancer Immunity Peptide Database48, across a panel of 28 human tissues for which RNA-Seq data were publicly available (see Table 6). Expression values ​​were normalized to rphm (see MCS Peripheral Expression section in Example 1 below for details) and averaged across all available RNA-Seq experiments for each tissue. For each antigen, the number of tissues in which its MCS is expressed > 15 rphm is shown on the left side of the heat map. Adip.sub.: subcutaneous adipocytes. Figure 6C: Most human TSAs are derived from non-coding regions. Bar graph depicting the number of human TSAs derived from in-frame translation of coding exons (in-frame), out-of-frame translation of coding exons (out-of-sense), and translation of putative non-coding (non-coding) regions. . The characteristics of the human TSAs identified in each sample can be found in Tables 2a-d and 3a-c. Figures 7A-D are architectural schematics of the codes used for the k-mer profiling workflow. Details related to the codes used to generate k-mers from RNA-seq reads (Figure 7A), filter k-mers (Figure 7B), assemble k-mers into contigs (Figure 7C) and translate contigs (Figure 7D). Figures 8A-C show the TSA validation process. Figure 8A: Schematic detailing the calculation of immunogenic status for MAP / protein pairs. FC: tumoral / syngeneic mTEChl (murine samples) or TEC / mTEC (human samples). Figure 8B: Strategy used to perform MS-related validations of MAPs marked as TSA candidates. Figure 8C: Schematic summarizing the strategy used to assign a gene location to MS-validated murine TSA candidates (CT26 and EL4), as well as MS-validated human TSA candidates for B-ALL and lung cancer samples. Figures 9A-D are graphs showing detection of antigen-specific CD8+ T cells in naïve and preimmunized mice. Figure 9A: Activation strategy for detection of tetramer+CD8+ T cells from pMHC ex vivo. Tetramer enrichment was performed on single cell suspensions isolated from the spleen and lymph nodes of each mouse. After doublet exclusion, Dump ’CD3+ cells were analyzed for CD8 and CD4 expression and pMHC I tetramer+ cells were analyzed in the CD8+ compartment. A representative staining obtained after enriched VTPV / H-2Kb-PE and M45 / H-2Db-APC tetramers in a naïve mouse is shown. Absolute numbers of tetramer+CD8+ T cells detected for each specificity are indicated. The Dump channel corresponds to positive pooled events for dead cells, CD45R and CD19, F4 / 80, CDllb, CDllc. Figures 9B, C: Representative analysis of CD44 expression on antigen-specific CD8 T cells in naïve (top row) and preimmunized (bottom row) mice. CD44 status of CD8+ cells before magnetic enrichment (FIG. 9B, left panel) and after ex vivo enrichment for viral specificities (FIG. 9B) and tetramer+ TSA specificities (FIG. 9C) are depicted. The percentages and number of CD44 positive or negative cells are indicated. Figure 9D: A representative experiment of the frequency of IFN-γ secreting CD8+ T cells in immunized and naïve mice. The number of dot forming units (SFU) relative to the number of CD8+ T cells in each condition are indicated below each well. Figures 10A-D are graphs showing the frequencies of antigen-specific T cells. Figure 10A: Frequencies of antigen-specific T cells in mice naïve or immunized with relevant or irrelevant peptides. Figures 10B, C: Frequencies of antigen-specific CD8+ T cells in mice immunized against VTPVYQHL or TVPLNHNTL (Figure 10B) or against VNYLHRNV or VNYIHRNV (Figure 10C) that were retested with EL4 cells at day 150. For the purposes of For comparison, the frequencies of antigen-specific T cells in naïve and immunized mice reported in Figure 10A are reproduced. Figure 10D: Frequencies of antigen-specific T cells in non-immunized mice injected with EL4 cells. All calculated frequencies of CD8+ T cell tetramers+ are expressed as the number of antigen-specific CD8+ T cells per 106 CD8+ T cells. Each symbol represents a mouse (n=1 to 9 mice). The dotted line represents a minimum level of detection of a T cell tetramer+ by 106 CD8+ T cells. Viral peptides used as controls are highlighted in grey. P values ​​were calculated using two-tailed Mann-Whitney tests (*p < 0.05). Figures 11A-C are graphs showing the correlation between antigen-specific T cell frequencies in naïve and preimmunized mice. Correlation between the frequencies of antigen-specific CD8+ T cells in the naïve repertoire and in immunized mice calculated by tetramer staining (Figure 11A) and IFN-γ ELISpot assays (Figure 11B). Figure HC: Correlation between the frequencies of antigen-specific TCD8+cccznn / i znz / B / v cells in immunized mice calculated by tetramer staining and IFN-γ ELISpot assays. Average frequencies were used to plot the data. The suitability of the curves was determined by the coefficient of determination (r2). Figures 12A-B represent an overview of the human TEC and mTEC transcriptomic landscapes. Figure 12A: Human TEC (062015 and 102015) and mTEC (S5 to Sil) isolated from unrelated donors show similar transcriptomic profiles. After RNA-Seq, transcripts expressed in at least one donor with a tpm > 1 were selected, as estimated by kallisto, to represent all one-to-one scatterplots. The Spearman rank correlation coefficient () is indicated in the upper left corner of each graph and the black line represents the identical expression of the transcripts. Figure 12B: RNA sequencing of additional human ECT / mTEC samples should result in minimal information gain. Using the set of expressed transcripts (tpm > 1 in at least one sample), the cumulative number of transcripts (cT) to be detected by adding additional samples to the cohorts (nS, see section Cumulative number of transcripts detected in TEC samples). and mTEC from Example 1 below) was extrapolated to x (nS - 1) cT =---------- + c by using the following function: + (nS-Ι)] with a = 23,892, 73, second = 0.8243389 and c = 75,976.11 (gray line). The graph indicates the cumulative number of transcripts detected when testing nS = 6 (the current cohort, black dots) or nS = 20 samples, as well as the total number of transcripts to be detected, which corresponds to fax (nS - 1) \ lim x í ς_ ni+ c=G+c= 99,868 ns^oo \L (n )J / (asymptote value). Figures 13A-C are graphs showing the activation strategies of cells isolated by FACS sorting. Figure 13A: Activation strategy for the isolation of mTEChlmurins. The isolation of mTECh was performed on single cell suspensions isolated from the thymus of C57BL / 6 or Balb / c mice. After doublet exclusion, mTEChlas cells were defined as 7-AAD', EpCAM+, CD45' (Alexa Fluor 700 for C57BL / 6 or FITC for Balb / c mice), UEA-1+, and LAb+ (C57BL / 6 mice). or IA / IE+(Balb / c mice). Figure 13B: Activation strategy for the isolation of human TECs and mTECs. Cell sorting was performed on single cell suspensions isolated from thymus that were obtained from individuals 3 months to 7 years of age undergoing corrective cardiovascular surgery. After exclusion of doublets, TECs were defined as CD45', 7-AAD', EpCAM+, and HLA-DR+. For mTEC classification, all 17 cells were further defined as CDR2-. Figure 13C: Gating strategy for the isolation of CD8+ T cells for IFN-γ ELISpot assays. Isolation of CD8+ T cells was performed on single cell suspensions isolated from the spleen of naïve or immunized C57BL / 6 mice. After doublet exclusion, the CD8a marker was used to enrich for CD8+ T cells. DESCRIPTION OF THE INVENTION The terms and symbols of genetics, molecular biology, biochemistry, and nucleic acids used in the present description follow those of treatises and standard texts in the field, for example Kornberg and Baker, DNA Replication, Second Edition (W.H. Freeman & Co, New York, 1992)', Lehninger, Biochemistry, Sixth Edition (W.H. Freeman & Co, New York, 2012); Strachan and Read, Human Molecular Genetics, Fifth Edition (CRC Press, 2018; Eckstein, editor, Oligonucleotides and Analogs: A Practical Approach (Oxford University Press, New York, 1991); and the like. All terms are to be understood with their typical meanings established in the corresponding technique. The articles "a" and "an" are used in the present description to refer to one or more than one (ie, to at least one) of the grammatical object of the article. By way of example, an "element" means one element or more than one element. Throughout this description, unless the context otherwise requires, the words comprehend, comprises, and comprising are understood to imply the inclusion of a stated stage or element or group of stages or elements but not the inclusion of it. exclusion of any other stage or element or group of stages or elements. The list of ranges of values ​​in this description is only intended to serve as a quick writing method for referring individually to each separate value within the range, unless otherwise indicated in this description, and each separate value is incorporated into the description as if indicated individually in the present description. All subsets of values ​​within the ranges are further incorporated into the description as if they were individually mentioned herein. All of the methods described in the present description may be carried out in any suitable order unless otherwise indicated in the present description or otherwise clearly contradicted by the context. The use of any and all examples or illustrative language (for example, "such as") provided herein is merely for the purpose of further illustrating the invention and is not intended to be a limitation on the scope of the invention unless otherwise noted. claim in any other way. No language in the description should be construed as indicating any element not claimed as essential to the practice of the invention. In the present description, the term approximately has its usual meaning. The term "approximately" is used to indicate that a value includes error variation inherent to the device or method used to determine the value, or encompasses values ​​close to the listed values, for example within 10% or 5 % of the values ​​mentioned (or range of values). Considerable effort is being devoted to discovering processable TSAs that can be used in therapeutic cancer vaccines. The most common strategy relies on reverse immunology: i) exorn sequencing is performed on tumor cells to identify mutations, and ii) MHC binding prediction software tools are used to identify which mutated MAPs might be good MHC1112 binders. . Although reverse immunology can enrich TSA candidates, at least 90% of these candidates are false positives5,13 because available computational methods can predict MHC binding, but cannot predict other steps involved in MAP processing14, fifteen. To overcome this limitation, some studies include mass spectrometry (MS) analyzes in their TSA16 discovery pipeline, thereby providing a rigorous molecular definition of various TSAs17,18. However, the performance of these approaches has been extremely poor: in melanoma, one of the most mutated tumor types, MS has validated an average of 2 TSAs per individual tumor19, while only a handful of TSAs have been found for other types of cancer15. The paucity of TSA is puzzling because injection of TILs or immune checkpoint inhibitors would not cause tumor regression if tumors did not express immunogenic antigens20. It was assumed that exonic mutation-based approaches had failed to identify TSAs because they did not take into account two crucial elements. First, these approaches focus solely on mTSAs and neglect aeTSAs, essentially because there is currently no method for high-throughput identification of aeTSAs. This represents a major deficiency because, although mTSAs are private antigens, cccznn / i znz / B / v aeTSAs would be preferred targets for vaccine development, as they can be shared by multiple tumors.7,9. Second, focusing on exorna as the only source of UT is too restrictive. The exorna (ie all protein-coding genes) represents only 2% of the human genome, while up to 75% of the genome can be transcribed and potentially translated22. In the studies described herein, the present inventors have developed a proteogenomic workflow capable of identifying non-tolerogenic TSAs, whether derived from coding or non-coding regions, simple or complex rearrangements, or simply cancer-restricted EREs. To identify non-tolerogenic sequences, rather than attempting to map all RNA sequencing reads and reconstruct potential mutations present there, the normal-matched correct signal, i.e., that of mTECh, was subtracted from the cancer signal and used. in silico translation of the resulting sequences as a database for MS. Compared to other techniques, the k-mer profiling workflow described in the present description has several advantages: (i) It is fast. Generation of the k-mer derived portion of the augmented cancer database typically takes less than half a day. (ii) It is impartial. It captures all cancer-specific sequences regardless of their nature, as demonstrated by the identification of TSAs derived from non-coding regions, as well as a TSA derived from a ~7500 bp deletion, (iii) It is modular. To enrich for non-tolerogenic sequences, cancer data was filtered in mTEChl, which proved to be a good proxy for peripheral expression of antigens, but data can be filtered in any other way, for example by removing all ENCODE data or adding dbSNP To the mix. The associated k-mer database is generated and added to the collection of normal samples for filtering. In one aspect, the present disclosure provides a method for identifying a candidate tumor antigen in a sample of tumor cells, the method comprising: (a) Generating a database of tumor-specific proteomes by: (i) extracting a set of subsequences (k-mers) comprising at least 33 base pairs of tumor RNA sequences (eg, RNA sequences obtained by whole transcriptome sequencing of the tumor cell sample); (ii) comparing the set of tumor subsequences from (i) with a set of corresponding control subsequences comprising at least 33 base pairs extracted from normal cell RNA sequences; (iii) extracting the tumor subsequences that are absent, or underexpressed by at least 4-fold, in the corresponding control subsequences, to thus obtain tumor-specific subsequences; and (iv) translating the tumor-specific subsequences in silico, to thereby obtain the tumor-specific proteome database; (b) generating a personalized database of the tumor proteome by: (i) comparing the tumor RNA sequences with a reference genome sequence to identify single base mutations in said tumor RNA sequences; (ii) inserting the single base mutations identified in (i) into the reference genome sequence, thereby creating a custom genome sequence; (iii) translating in silico the transcripts encoding expressed proteins from said custom genome sequence, to thereby obtain the custom proteome database; (c) comparing the sequences of the major histocompatibility complex (MHC)-associated peptides (MAP) of said tumor with the sequences of the tumor-specific proteome database of (a) and the customized tumor proteome database of (b) to identify MAPs; and (d) identifying a candidate tumor antigen among the MAPs identified in (c), wherein a candidate tumor antigen is a peptide whose sequence and / or coding sequence is overexpressed in tumor cells relative to normal cells. In one embodiment, the above method further comprises isolating and sequencing major histocompatibility complex (MHC)-associated peptides (MAP) from the tumor cell sample. In one embodiment, the above method further comprises performing whole transcriptome sequencing on the tumor cell sample, thereby obtaining tumor RNA sequences. The term candidate tumor antigen, as used herein, refers to a peptide that binds to a major histocompatibility molecule (MHC) and is present on the surface of tumor cells only, or present at significantly higher levels / frequencies. high (at least 2 times, preferably at least 4, 5 or 10 times) on the surface of tumor cells relative to non-tumor cells. Said candidate tumor antigen can be targeted to induce a T cell response against tumor cells expressing the antigen on their surface. Methods for isolating MHC-associated peptides (MAPs) from a cell sample are well known in the art. The most commonly used technique is mild acid elution (MAE) of MHC-associated peptides from living cells, as described in Fortier et al. (J. Exp. Medicina. 205(3): 595-610, 2008). Another technique is immunoprecipitation or affinity purification of peptideMHC class I complexes followed by peptide elution (see, eg, Gebreselassie et al., Hum Immunol. 2006 Nov;67(11):894-906). Two high-yield strategies based on the latter approach have been implemented. The first is based on the transfection of cell lines with expression vectors encoding soluble secreted MHC (lacking a functional transmembrane domain) and the elution of secreted MHC-associated peptides (Barnea et al., Eur J Immunol. Jan 2002; 32(1):213-22;and Hickman HD et al., J Immunol. 2004 Mar 1;172(5):2944-52). The second approach relies on chemical or metabolic labeling to provide quantitative profiles of MHC-associated peptides (Weinzieri AO et al, Mol Cell Proteomics. 2007 Jan;6(1):102-13. Epub 2006 Oct 29; Lemmel C et al, Nat Biotechnol 2004 Apr;22(4):450-4 Epub 2004 Mar 7 Milner E, Mol Cell Proteomics 2006 Feb;5(2):357-65 Epub 4 Nov of 2005). Eluted MAPs can be subjected to any purification / enrichment step, including size exclusion chromatography or ultrafiltration (using a filter with a cut-off of about 5000 Da, e.g. about 3000 Da), reverse phase chromatography ( hydrophobic) and / or ion exchange chromatography (e.g., cation exchange chromatography), prior to further analysis. The sequence of the eluted MAPs can be determined using any method known in the art for sequencing peptides / proteins, such as mass spectroscopy (tandem mass spectrometry or MS / MS, as described below and the degradation reaction of Edman. Whole transcriptome sequencing (also called total RNA sequencing, RNA sequencing, or RNA-seq) refers to the sequencing of all RNAs present in a sample (tumor sample, normal cell sample), including coding RNAs as well as well as multiple forms of non-coding RNA such as miRNA, snRNA, and tRNA. Methods for performing complete transcriptome sequencing, eg, next generation sequencing (NGS) methods, are well known in the art. Various NGS platforms that are commercially available (e.g. from Illumina (NextSeq™, HiSeq™), Thermofisher (Ion Total™ RNA-Seq Kit), Clontech (SMARTer™) or that are mentioned in the literature can be used in the described method. in the present description, for example those described in detail in Zhang et al. 2011: The impact of next-generation sequencing on genomics. J. Genet Genomics 38(3), 95-109; or in Voelkerding et al. 2009: Next Generation sequencing: From basic research to diagnostics.Clinical Chemistry 55, 641-658. Preferably, the RNA preparations serve as starting material for NGS. Such nucleic acids can be readily obtained from samples such as biological material, for example from formalin-fixed, fresh, snap-frozen or paraffin-embedded tumor tissues or from freshly isolated cells or from circulating tumor cells (CTCs) that are present in the peripheral blood of patients. patients. Normal or control RNAs can be extracted from normal somatic tissue or germ line cells. RNA sequences from normal cells may correspond to a collection of RNA sequences from different types of normal cells, eg normal cells from different tissues. RNA sequences from normal cells can also be obtained from thymic cells, preferably medullary thymic epithelial cells (mTEC) such as MHC II medullary thymic epithelial cells (mTECh). mTEChl cells advantageously have a unique promiscuous gene expression profile, as they express approximately 70-90% of somatic cell protein-coding sequences, and their MAPs can induce central immune tolerance. The method described in the present disclosure comprises generating a database of tumor-specific proteomes by using an alignment-free RNA sequence analysis workflow, termed k-mer profiling, comprising sequences derived from translation of variants. structural (any type of mutations, including large insertions or deletions (InDels) or fusions) and non-coding regions. Tumor and normal RNA sequences (RNA-seq reads) are cut or divided into k-mers, ie, subsequences of length k with k > 33 nucleotides. Since peptides bound to MHC class I (MAP) molecules are generally no more than 11 amino acids in length (and thus encoded by 33 nucleotide sequences), divide RNA sequences into subsequences of at least 33 nucleotides cccznn / i znz / B / v minimizes the risk of missing potential MAPs. The skilled person will understand that to minimize the size of the tumor-specific proteome database, it is preferred to divide RNA sequences into subsequences of 33 nucleotides (i.e., k = 33 nucleotides) to identify MHC class I-restricted tumor antigens. The skilled person will also understand that to identify MHC class II restricted tumor antigens, the minimum k-mer length should be increased from 33 to 54 nucleotides (k > 54 nucleotides), MHC II associated peptides generally range from 13 to 18 amino acids long. Tumor subsequences are then compared to a set of corresponding control subsequences (from RNA sequences from normal cells) to extract tumor subsequences that are absent or underexpressed by at least 4-fold (preferably at least 5, 6, 7, 8, 9 or 10 times), in the corresponding control subsequences. In one embodiment, to minimize the redundancy inherent in the k-mer space, the method further comprises assembling overlapping tumor-specific subsequences into longer tumor subsequences (typically referred to as contigs). The tumor-specific subsequences or contigs are then translated in silico (for example, frame 3 or frame 6 translated, depending on whether the subsequences or contigs are derived from the sense or non-sense strand) to obtain the database of specific proteomes. of the tumor. In one embodiment, protein fragments less than 8 amino acids (the minimum length of MHC class I peptides) or 13 amino acids (the minimum length of MHC class II peptides) are removed from the tumor-specific proteome database. . In one embodiment, the method further comprises generating a k-mer database with k = 24 nucleotides (for MHC class I peptides) or k = 39 (for MHC class II peptides) from the RNA sequences (from normal cells and tumor) to obtain cancer / tumor and normal k-mer databases of 24 (or 39) nucleotides in length. These databases can be used to compare with MAP coding sequences (MCS) to determine if MCS are overexpressed or overrepresented in tumor cells, as described below. The method further comprises the generation of a personalized database of the tumor proteome. To do this, the tumor RNA sequences (tumor RNA-seq reads) are compared to a reference genome sequence to identify single base mutations in the tumor RNA sequences. These mutations are then inserted into the reference genome to obtain a custom tumor genome, from which it is possible to obtain the corresponding custom tumor proteome database containing the canonical translation product sequences of all transcription sequences. that encode expressed proteins. The generation of a customized tumor proteome database, which allows the identification of mutated WT MAP and TSA (neoantigens) encoded by the exorna canonical framework, also improves the reliability of databases used for MS analysis by not overly biasing the database towards tumor-specific sequences, which would lead to the identification of several false positives. In one embodiment, the method further comprises generating a custom normal proteome database. To do this, RNA sequences from normal cells (normal RNA sequence reads) are compared to a reference genome sequence to identify single base mutations in normal RNA sequences. These mutations are then inserted into the reference genome to obtain a custom normal genome, from which it is possible to derive the corresponding custom normal proteome database containing the canonical translation product sequences of all transcription sequences. that encode expressed proteins. This custom database of normal proteomes can be used to filter MAPs expressed in normal (non-tumor) cells, which are not suitable candidates for TSA. The term reference genome, as used herein, refers to human genome assemblies described in the literature, and includes, for example, Genome Reference Consortium Human Build 38 (GRCh38, RefSeq: accession number GCF_000001405.37). , Hs_Celera_WGSA (Celera Genomics; Istrail S. et al., Proc Nati Acad Sci USA 2004;101(7):1916-21). Epub 9 Feb 2004), HuRef and HuRefPrime (J. Craig Venter Institute; Levy S, et al. PLoS Biology. 2007; 5: 2113-2144), YH1 and BGIAF (Beijing Genomics Institute; Li R, et al. Genome Research 2010;20:265-272), HsapALLPATHS 1 (Broad Institute) and the like. A list of reference human genome assemblies can be found in the National Center for Biotechnology Information (NCBI) Assembly database. In one embodiment, the reference genome is GRCh38. The sequences of the MAPs obtained in step (a) of the method are then compared with (eg, attacked against) the sequences of the tumor-specific proteome database and the personalized tumor proteome database, which allows the MAP identification. Candidate tumor antigens can be identified among the previously identified MAPs. Said candidate tumor antigens correspond to peptides whose sequences and / or coding sequences are overexpressed in tumor cells relative to normal cells. In one embodiment, the method further comprises deleting or discarding MAPs whose sequences are detected in the normal custom proteome database. In one embodiment, the method comprises retrieving the coding sequences of the identified MAPs, ie, the MAP Coding Sequence (MCS). In another embodiment, the method comprises transforming the MCS into k-mer pools of 24 (for MHC class I peptides) or 39 (for MHC class II peptides) nucleotides. In another embodiment, these MCS-derived k-mer pools are compared to cancer / tumor and normal 24 (or 39) nucleotide k-mer databases. In one embodiment, the method comprises: (a) isolating and sequencing major histocompatibility complex (MHC)-associated peptides (MAP) in a sample of tumor cells; (b) performing whole transcriptome sequencing on said sample of tumor cells, to thereby obtain tumor RNA sequences; (c) generating a database of tumor-specific proteomes by: (i) extracting a set of subsequences (k-mers) comprising at least 33 nucleotides from said tumor RNA sequences; (ii) comparing the set of tumor subsequences from (i) with a set of corresponding control subsequences comprising at least 33 nucleotides extracted from normal cell RNA sequences; (iii) extracting the tumor subsequences that are absent, or underexpressed by at least 4-fold, in the corresponding control subsequences, to thus obtain tumor-specific subsequences; and (iv) translating the tumor-specific subsequences in silico, to thereby obtain the tumor-specific proteome database; (d) generating a personalized database of the tumor proteome by: (i) comparing the tumor RNA sequences with a reference genome sequence to identify single base mutations in said tumor RNA sequences; (ii) inserting the single base mutations identified in (i) into the reference genome sequence, thereby creating a personalized tumor genome sequence; (iii) translating in silico the transcripts encoding expressed proteins from said personalized tumor genome sequence, to thereby obtain the personalized tumor proteome database; (e) generate a custom normal proteome database by: (i) comparing RNA sequences from normal cells to a reference genome sequence to identify single base mutations in said normal RNA sequences; (ii) inserting the single base mutations identified in (i) into the reference genome sequence, thereby creating a custom normal genome sequence; (iii) translating in silico the expressed protein-encoding transcripts from said custom normal genome sequence, to thereby obtain the custom normal proteome database; (f) generating a normal and tumor k-mer database (i) extracting a set of subsequences comprising at least 24 nucleotides from said normal cell RNA sequences and said tumor RNA sequences; (g) comparing the sequences of the MAPs obtained in (a) with the sequences of the tumor-specific proteome database of (c) and the customized tumor proteome database of (d) to identify the MAPs; and (h) identifying a candidate tumor antigen among the MAPs identified in (f), wherein a candidate tumor antigen corresponds to a MAP (1) whose sequence is not present in the custom normal proteome database; and (2) (i) whose sequence is present in the personalized tumor proteome database; and / or (i) whose coding sequence is overexpressed or overrepresented in said tumor k-mer database relative to said normal kmer database. In one embodiment, the coding sequence is transformed into a set of MAP-derived k-mers (eg, 24 nts k-mers), and the expression or representation of the MAP-derived k-mers in the databases is determined. k-mer tumor and normal data. Overexpressed or overrepresented as used herein means that the sequence is present in the tumor k-mer database at a level that is at least 2-fold, more preferably 3, 4, or 5-fold, and most preferably at least 10 times, relative to the normal k-mer database. In one embodiment, the MAP-derived k-mer or coding sequence is absent from the normal k-mer database. In one embodiment, referring to Figure 7A, identification and validation of the TSA candidate is accomplished in the following manner. Each MAP and its associated MAP coding sequences (MCS) is queried in the relevant normal or cancer custom proteome and the 24 nucleotide long normal and cancer k-mer databases. MAPs detected in the normal custom proteome were excluded. Only MAPs present in the cancer personalized proteome and / or cancer k-mer database are identified / selected as TSA candidates. For MAPs absent from both custom proteomes but present in both kmer databases, they are selected if their MCS is overexpressed (eg, at least 2-fold, more preferably at least 5-fold, and most preferably at least 10-fold) in cancer cells relative to normal cells. If the MAP is encoded by multiple MCSs, it is identified / selected as a TSA candidate if their respective MCSs were concordant, ie if it is consistently marked as a TSA candidate. In one embodiment, since they are difficult to distinguish by MS, TSA candidates with PL variants are excluded as TSA candidates. In one embodiment, prior to comparison, eluted MAPs are filtered to select for peptides 8 to 11 amino acids in length. In another embodiment, prior to comparison, the eluted MAPs are filtered to select those that have a percentile rank < 2% for at least one of the relevant MHC I molecules, as predicted by the NetMHC version 4.0 software (httD: / / www. cbs.dtu.dk / services / NetMHC-4.0) (Andreatta M, Nielsen M, Bioinformatics 2016 Feb 15;32(4):511-7; Nielsen M, et al., Protein Sci., (2003)12: 1007-17). In one embodiment, the method further comprises comparing the candidate tumor antigen coding sequence with sequences from normal tissues. In modalities, the sequences of at least 5, 10, 15, 20 or 25 different tissues are used. Normal tissue sequences can be obtained from public databases, such as Expression Atlas (Petryszak et al., Nucleic Acids Research, Volume 44, Issue DI, Jan 4, 2016, pages D746-D752), scRNASeqDB (Cao Y, et al. (2017).Gene58(12), 368), RNA-Seq Atlas (Krupp et al, Bioinformatics, Volume 28, Issue 8, 2012 Apr 15, pages 1184-1185) and Encode, or can be generated by performing RNAseq on normal tissues. In one embodiment, the method further comprises selecting the candidate tumor antigen if (1) its coding sequence is not expressed in any of the normal tissues tested, or if it is expressed only in MHC class I negative tissues, or (2) its coding sequence that is expressed is less than 50%, preferably less than 45, 40, 35 or 30% of the MHC class I positive tissues tested. In one embodiment, the candidate tumor antigen is selected if its coding sequence is expressed in fewer than 7, preferably fewer than 6, 5, 4, or 3 of the normal tissues tested. In one embodiment, the method further comprises determining the genomic location of the TSA candidate coding sequence and selecting the TSA candidate if (1) the coding sequence matches a matched genomic location; (2) the coding sequence does not match a hypervariable region (such as H2 genes, TCR Ig) or multiple genes; and (3) do not overlap with synonymous mutations. Said determination can be made using the BLAT tool of the UCSC Genome Browser (Kent WJ. Genome Res. April 2002;12(4): 656-64) and / or the Integrative genomics viewer (IGV) (IGV) tool (Robinson et al.. 2011 Nat Biotechnol Jan;29(1):24-6). In one embodiment, the method further comprises determining or predicting the binding of the identified tumor candidate antigen (TSA candidate) to an MHC class I molecule. The binding may be a predicted binding affinity (IC50) of peptides to the allelic products, which can be obtained by using tools such as NetMHC. An overview of the various MHC class I peptide binding tools available is provided in Peters B et al., PLoS Comput Biol 2006, 2(6):e65; Trost et al., Immunoma Res 2007, 3(1):5; Lin et al., BMC Immunology 2008, 9:8). Binding of the identified TSA candidate to an MHC class I molecule can be determined using other known methods, eg, the T2 Peptide Binding Assay. T2 cell lines are deficient in TAP but still express low amounts of MHC class I on the cell surface. The T2 binding assay is based on the ability of peptides to stabilize the MHC class I complex on the surface of the T2 cell line. T2 cells are incubated with a specific peptide (for example, a TSA candidate), stabilized MHC class I complexes are detected using a pan-HLA class I antibody, analysis is performed (by cytometry for example) and binding is assessed relative to a non-binding negative control. The presence of stabilized peptide / MHC class I complexes on the surface is indicative that the peptide (eg, TSA candidate) binds to MHC class I molecules. The binding of a peptide of interest (eg, TSA candidate) to MHC can also be assessed based on its ability to inhibit the binding of a radiolabeled probe peptide to MHC molecules. MHC molecules are solubilized with detergents and purified by affinity chromatography. They are then incubated for 2 days at room temperature with the peptide of interest (eg, TSA candidate) and an excess of a radiolabeled probe peptide, in the presence of a cocktail of protease inhibitors. At the end of the incubation period, the MHC-peptide complexes are separated from unbound radiolabeled peptide by size exclusion gel filtration chromatography, and the percent bound radioactivity is determined. The binding affinity of a particular peptide for an MHC molecule can be determined by co-incubating various doses of the unlabeled competitor peptide with the MHC molecules and the labeled probe peptide. The concentration of unlabeled peptide required to inhibit binding of the labeled peptide by 50% (IC50) can be determined by plotting dose versus % inhibition (see, for example, Current Protocols in Immunology (1998) 18.3.1-18.3. 19, John Wiley & Sons, Inc.). Binding of the identified TSA candidate to an MHC class I molecule can also be determined using a T cell epitope discovery system / tool, such as the Prolmmune REVEAL® and Pro VE® T cell epitope discovery systems or the NetMHC tool (see, for example, Desai and Kulkarni-Kale, Methods Mol Biol. 2014; 1184: 333-64). In one embodiment, the method further comprises assessing the number or frequency of T cells that recognize the candidate tumor antigen in a cell population, eg, in a sample of cells (eg, PBMC) from a subject. The number or frequency of T cells that recognize a given antigen can be assessed using various methods known in the art, for example, by contacting the cell population; with multimeric MHC class I molecules (for example, MHC tetramers ) comprising said candidate tumor antigen in their peptide-binding groove, and determining the number of cells labeled with the multimeric MHC class I molecules. Multimeric MHC class I molecules can be detectably labeled with a fluorophore (direct labeling) or they can be labeled with a moiety that is recognized by a labeled ligand (indirect or secondary labelling). Alternatively, the number or frequency of T cells recognizing the TSA candidate can be assessed by determining the number / frequency of activated T cells in the presence of the TSA candidate under conditions suitable for T cell activation. The number / frequency of T cells Activated cells cccznn / i znz / B / v can be assessed by detecting cells that secrete a cytokine induced by T cell activation, eg IFN-γ or IL-2 (eg by ELISpot or flow cytometry). In one embodiment, the method further comprises assessing the ability of the candidate tumor antigen to induce T cell activation, eg, by contacting a population of T cells with cells (eg, APCs such as dendritic cells) bearing the antigen. candidate tumor bound to MHC class I molecules on its cell surface, and that measure at least one parameter of T cell activation, such as proliferation, cytokine / chemokine production (e.g., IFN-γ or IL- 2), cytotoxic destruction and the like. In one embodiment, the method further comprises evaluating the ability of the candidate tumor antigen to kill T-cell mediated tumor cells and / or to inhibit tumor growth. This can be achieved by using tumor cells in vitro, or by using a suitable animal model in vivo. In one embodiment, the candidate tumor antigen is about 7 to 20 amino acids in length, and more particularly about 8 to 18 amino acids in length, preferably 8 to 11 (for MHC class I tumor antigens) or 13 to 18 ( for MHC class I) amino acids in length. MHC class II tumor antigens) amino acids. The methods described herein may be useful to identify candidate tumor antigen for any type of cancer by performing whole transcriptome sequencing on the tumor / cancer cell sample of interest. Examples of such cancers include, but are not limited to, carcinoma, lymphoma, blastema, sarcoma, and leukemia, and more particularly bone cancer, blood / lymphoid cancer such as leukemia (AML, CML, ALL), myeloma, lymphoma, lung cancer. , liver cancer, pancreatic cancer, skin cancer, head or neck cancer, cutaneous or intraocular melanoma, uterine cancer, ovarian cancer, rectal cancer, cancer of the anal region, stomach cancer, colon cancer, cancer cancer, prostate cancer, uterine cancer, carcinoma of the sexual and reproductive organs, cancer of the esophagus, cancer of the small intestine, cancer of the endocrine system, cancer of the thyroid gland, cancer of the parathyroid gland, cancer of the adrenal gland, soft tissue sarcoma, bladder cancer, kidney cancer, renal cell carcinoma, renal pelvic carcinoma, central nervous system (CNS) neoplasms, neuroectodermal cancer, spinal axis tumors, glioma, meningioma, and pituitary adenoma. Thus, in one embodiment, the tumor cell sample used in step (a) of the method described herein is a sample comprising cells from any of the cancers listed above. In another aspect, the present description refers to a tumor antigenic peptide (or tumor-specific peptide) identified in the present description, that is, comprising one of the amino acid sequences described in Tables la, Ib, 2a-2d or 3a -3c (SEQ ID NO: 1-39), preferably Tables 2a-2d or 3a-3c (SEQ ID NO: 17-39), or a variant thereof having one or more mutations with respect to the sequences of SEQ ID NO: 1-39. In general, peptides such as tumor antigen peptides presented in the context of HLA class I range in length from about 7 or 8 to about 15, or preferably 8 to 14 amino acid residues. In some embodiments of the methods of the disclosure, the longer peptides comprising the tumor antigen peptide sequences defined herein are artificially loaded into cells such as antigen presenting cells (APCs), processed by the cells, and the peptide Tumor antigens are presented by MHC class I molecules on the surface of the APC. In this method, peptides / polypeptides of more than 15 amino acid residues (i.e., a tumor antigen precursor peptide) can be loaded into the APC, processed by proteases in the cytosol of the APC to yield the corresponding tumor antigen peptide as described. defined in the present description for presentation. In some embodiments, the precursor peptide / polypeptide that is used to generate the defined tumor antigen peptide in some embodiments is, for example, 1000, 500, 400, 300, 200, 150, 100, 75, 50, 45, 40, 35 , 30, 25, 20 or 15 amino acids or less. Therefore, all methods and processes using the tumor antigen peptides described herein include the use of longer peptides or polypeptides (including native protein), i.e., tumor antigen precursor peptides / polypeptides, to induce tumor antigen peptides. final 8-14 tumor antigen peptide presentation after processing by the cell (APC). In some embodiments, the tumor antigen peptide mentioned herein is about 8 to 14, 8 to 13, or 8 to 12 amino acids in length (eg, 8, 9, 10, 11, 12, or 13 amino acids in length), which small enough for a direct fit into an HLA class I molecule. In one embodiment, the tumor antigen peptide comprises 20 or fewer amino acids, preferably 15 or fewer amino acids, more preferably 14 or fewer amino acids. In one embodiment, the tumor antigen peptide comprises at least 7 amino acids, preferably at least 8 amino acids, more preferably at least 9 amino acids. The term amino acid, as used herein, includes L and D isomers of naturally occurring amino acids, as well as other amino acids (for example, naturally occurring amino acids, non-naturally occurring amino acids, amino acids that are not encoded by sequences). of nucleic acid, etc.) used in peptide chemistry to prepare synthetic analogues of tumor antigen peptides. Examples of natural amino acids are glycine, alanine, valine, leucine, isoleucine, serine, threonine, etc. Other amino acids include, for example, non-genetically encoded forms of amino acids, as well as a conservative L-amino acid substitution. Naturally occurring non-genetically encoded amino acids include, for example, beta-alanine, 3-amino-propionic acid, 2,3-diaminopropionic acid, alpha-aminoisobutyric acid (Aib), 4-amino-butyric acid, A-methylglycine (sarcosine) , hydroxyproline, ornithine (for example, Lornithine), citrulline, / -butylalanine, í-butylglycine, A-methylisoleucine, phenylglycine, cyclohexylalanine, norleucine (Nle), norvaline, 2-naphthylalanine, pyridylalanine, 3-benzothienyl alanine, 4-chlorophenylalanine, 2-fluorophenylalanine, 3-fluorophenylalanine, 4-fluorophenylalanine, penicillamine, l,2,3,4-tetrahydro-isoquinoline-3-carboxylic acid, beta-2-thienylalanine, methionine sulfoxide, Lhomoarginine (Hoarg), N-acetyl lysine , 2-aminobutyric acid, 2-aminobutyric acid, 2,4,diaminobutyric acid (D- or L-), p-aminophenylalanine, A-methylvaline, homocysteine, homoserine (HoSer), cysteic acid, epsilon-amino hexanoic acid , delta-aminovaleric acid or 2,3-diaminobutyric acid (D- or L-), etc. These amino acids are well known in the art of biochemistry / peptide chemistry. In one embodiment, the tumor antigen peptide comprises only naturally occurring amino acids. In embodiments, the tumor antigen peptides described herein include peptide variants with altered sequences that contain functionally equivalent amino acid residue substitutions, relative to the sequences mentioned herein. For example, one or more amino acid residues within the sequence can be substituted with another amino acid of similar polarity (having similar physicochemical properties) that acts as a functional equivalent, resulting in a silent alteration. The substitution of an amino acid within the sequence may be selected from other members of the class to which the amino acid belongs. For example, positively charged (basic) amino acids include arginine, lysine, and histidine (as well as homoarginine and ornithine). Nonpolar (hydrophobic) amino acids include leucine, isoleucine, alanine, phenylalanine, valine, proline, tryptophan, and methionine. Neutral polar amino acids include serine, threonine, cysteine, tyrosine, asparagine, and glutamine. Negatively charged (acidic) amino acids include aspartic acid and glutamic acid. The amino acid glycine can be included in the family of nonpolar amino acids or in the family of polar uncharged (neutral) amino acids. Substitutions made within a family of amino acids are generally understood as conservative substitutions. The tumor antigen peptide mentioned in the present description may comprise all L-amino acids, all D-amino acids or a mixture of L- and D-amino acids. In one embodiment, the tumor antigen peptide mentioned herein comprises all L-amino acids. In one embodiment, in tumor antigen peptide sequences comprising one of the sequences set forth in SEQ ID NO: 1-39, amino acid residues that do not contribute substantially to T-cell receptor interactions can be modified by replacing them with other amino acids whose incorporation does not substantially affect T cell reactivity and does not abolish binding to the relevant MHC molecule. In one embodiment, the tumor antigen peptide variant is sequence optimized to enhance MHC binding, i.e., it comprises one or more mutations (eg 1, 2 or 3 mutations), eg amino acid substitutions, that enhance binding. to the MHC molecule. Tumor antigen peptide variant binding affinities can be assessed, for example, using MHC binding prediction tools such as NetMHC4.0; NetMHCpan4.0; and MHCflurry 1.2.0. Sequence-optimized variants of tumor antigen peptides may be considered, for example, if the prediction of binding affinity for a specific HLA is equivalent to, or preferably stronger than, the native tumor antigen peptide. Selected sequence-optimized target peptides can be screened for binding to specific HLAs in vitro using methods known in the art, for example using Prolmmune's REVEAL assay. The tumor antigen peptide may also be N- and / or C-terminally blocked or modified to prevent degradation, increase stability, affinity and / or absorption, and therefore the present disclosure provides a variant tumor antigen peptide having the formula Z ^X-Z2, where X are the sequences of the tumor antigen peptides established in SEQ ID NO: 1-39, preferably 17-39. In one embodiment, the amino terminal residue (ie, the free amino group at the N-terminus) of the tumor antigen peptide is modified (eg, for protection against degradation), eg, by covalent attachment of a residue / chemical group (Z1) Z1 can be a straight or branched chain alkyl group of one to eight carbons, or an acyl group (R-CO-), where R is a hydrophobic moiety (for example, acetyl, propionyl, butanyl, isopropionyl or isobutanyl) or an aroyl group (Ar-CO-), where Ar is an aryl group. In one embodiment, the acyl group is a Ci-Có or C3-C16 acyl group (linear or branched, saturated or unsaturated), in a further embodiment, a saturated Ci-Có acyl group (linear or branched) or a C3 acyl group. -C6 unsaturated (linear or branched), for example, an acetyl group (CH3-CO-, Ac). In one embodiment, Z1 is absent. The carboxy terminal residue (i.e., the free carboxy group at the C-terminus of the tumor antigen peptide) of the tumor antigen peptide can be modified (eg, for protection against degradation), for example, by covalent attachment of a residue / chemical group (Z2), for example by amidation (replacement of the OH group by an NH2 group), so in such a case Z2 is an NH2 group. In one embodiment, Z2 may be a hydroxamate group, a nitrile group, an amide group (primary, secondary or tertiary), an aliphatic amine of one to ten carbons such as methylamine, isobutylamine, isovalerylamine or cyclohexylamine, an aromatic amine or arylalkylamine such as such as aniline, naphthylamine, benzylamine, cinnamylamine or phenylethylamine, an alcohol or CH2OH. In one embodiment, Z2 is absent. In one embodiment, the tumor antigen peptide comprises one of the sequences described in SEQ ID NO: 1-39, preferably 17-39. In one embodiment, the tumor antigen peptide consists of one of the sequences described in SEQ ID NO: 1-39, preferably 17-39, ie, where Z1 and Z2 are absent. In one embodiment, the present disclosure provides a tumor antigen peptide that binds to an HLA molecule of the HLA-A2 allele, preferably the HLA-A*02:01 allele, and comprises or consists of one of the amino acid sequences set forth in any of SEQ ID NO: 1719, 27 and 28. In one embodiment, the present disclosure provides a tumor antigen peptide that binds to an HLA molecule of the HLA-B40 allele, preferably the HLA-B*40:01 allele, and comprises or consists of the amino acid sequence set forth in SEQ ID NO: 20. In one embodiment, the present disclosure provides a tumor antigen peptide that binds to an HLA molecule of the HLA-A11 allele, preferably the HLA-A* 11:01 allele, and comprises or consists of one of the amino acid sequences set forth in any of SEQ ID NO: 2123 and 29-35. In one embodiment, the present disclosure provides a tumor antigen peptide that binds to an HLA molecule of the HLA-B08 allele, preferably of the HLA-B*08:01 allele, and comprises or consists of the amino acid sequences set forth in SEQ ID NO: 24 or 25. In one embodiment, the present disclosure provides a tumor antigen peptide that binds to an HLA molecule of the HLA-B07 allele, preferably the HLA-B*07:02 allele, and comprises or consists of the amino acid sequence set forth in SEQ ID NO: 26 or 36. In one embodiment, the present disclosure provides a tumor antigen peptide that binds to an HLA molecule of the HLA-A24 allele, preferably the HLA-A*24:02 allele, and comprises or consists of the amino acid sequences set forth in SEQ ID NO: 38 or 39. In one embodiment, the present disclosure provides a tumor antigen peptide that binds to an HLA molecule of the HLA-C07 allele, preferably the HLA-C*07:01 allele, and comprises or consists of the amino acid sequence set forth in SEQ ID NO: 37. In one embodiment, the tumor antigen peptide is a leukemia tumor antigen peptide and comprises or consists of one of the amino acid sequences set forth in any of SEQ ID NO: 17-28. In one embodiment, the tumor antigen peptide is a lung tumor antigen peptide and comprises or consists of one of the amino acid sequences set forth in any of SEQ ID NO: 29-39. In one embodiment, the tumor antigen peptide is encoded by a sequence located in a non-coding region of the genome. In one embodiment, the tumor antigen peptide is encoded by a sequence located in a transcribed region (UTR), ie, a 3'-UTR or 5-UTR region. In another embodiment, the tumor antigen peptide is encoded by a sequence located in an intron. In another embodiment, the tumor antigen peptide is encoded by a sequence located in an intergenic region. In one embodiment, the tumor antigen peptide is encoded by a sequence located in an endogenous retroelement (ERE). In another embodiment, the tumor antigen peptide is encoded by a sequence located in an exon and originates from a reading frameshift. The tumor antigen peptides of the disclosure can be produced by expression in a host cell comprising a nucleic acid encoding the tumor antigen peptides (recombinant expression) or by chemical synthesis (eg, solid phase peptide synthesis). These peptides can be readily synthesized by manual and / or automated solid phase procedures well known in the art. Suitable syntheses can be performed, for example, using the T-boc or Fmoc procedures. Techniques and procedures for solid phase synthesis are described, for example, in Solid Phase Peptide Synthesis: A Practical Approach, by E. Atherton and RC Shcppard, published by IRL, Oxford University Press, 1989. Alternatively, tumor antigen peptides they can be prepared via segment condensation, as described, for example, in Liu et al., Tetrahedron Lett. 37:933-936, 1996; Baca et al., J. Am. Chem. Soc. 117: 1881-1887, 1995; Tam et al., Int. J. Peptide Protein Res. 45: 209-216, 1995; Schnolzer and Kent, Science 256: 221-225, 1992; Liu and Tam, J. Am. Chem. Soc. 116: 4149-4153, 1994; Liu and Tam, Proc. nati. Acad. Sci. USA 91: 65846588, 1994; and Yamashiro and Li, Int. J. Peptide Protein Res. 31: 322-334, 1988). Other useful methods for synthesizing tumor antigen peptides are described in Nakagawa et al., J. Am. Chem. Soc. 107: 7087-7092, 1985. In one embodiment, the tumor antigen peptide is chemically synthesized (synthetic peptide). Another embodiment of the present disclosure relates to a non-naturally occurring peptide wherein said peptide consists or consists essentially of an amino acid sequence defined in the present disclosure and has been synthetically produced (eg, synthesized) as a pharmaceutically acceptable salt. The salts of the tumor antigen peptides according to the present disclosure differ substantially from the peptides in their state(s) in vivo, since the peptides generated in vivo are not salted. The unnatural salt form of the peptide can modulate the solubility of the peptide, in particular in the context of pharmaceutical compositions comprising the peptides, for example peptide vaccines as described herein. Preferably, the salts are pharmaceutically acceptable salts of the peptides. In one embodiment, the tumor antigen peptide mentioned herein is substantially pure. A compound is substantially pure when it is separated from the components that naturally accompany it. Typically, a compound is substantially pure when it is at least 60%, more generally 75, 80%, or 85%, preferably greater than 90%, and more preferably greater than 95%, by weight, of the total material in a sample. Thus, for example, a polypeptide that is chemically synthesized or produced by recombinant technology will generally be substantially free of its naturally associated components, eg, components of its parent macromolecule. A nucleic acid molecule is substantially pure when it is not immediately contiguous with (ie, covalently linked to) the coding sequences with which it is normally contiguous in the natural genome of the organism from which the nucleic acid is derived. A substantially pure compound can be obtained, for example, by extraction from a natural source; by expression of a recombinant nucleic acid molecule encoding a peptide compound; or by chemical synthesis. Purity can be measured by using any appropriate method, such as column chromatography, gel electrophoresis, HPLC, etc. In one embodiment, the tumor antigen peptide is in solution. In another embodiment, the tumor antigen peptide is in solid form, eg, lyophilized. In another aspect, the disclosure further provides a nucleic acid (isolated) encoding the tumor antigen peptides mentioned herein or a tumor antigen precursor peptide. In one embodiment, the nucleic acid comprises from about 21 to about 45 nucleotides, from about 24 to about 45 nucleotides, eg, 24, 27, 30, 33, 36, 39, 42, or 45 nucleotides. Isolated, as used herein, refers to a peptide or nucleic molecule separated from other components that are present in the natural environment of the molecule or a naturally occurring macromolecule (eg, including other nucleic acids, proteins, lipids, sugars, etc.). Synthetic, as used herein, refers to a peptide or nucleic molecule that is not isolated from its natural sources, eg, produced by recombinant technology or through the use of chemical synthesis. A nucleic acid of the disclosure can be used for recombinant expression of the tumor antigen peptide of the disclosure, and can be included in a vector or plasmid, such as a cloning vector or an expression vector, that can be transfected into a host cell. In one embodiment, the disclosure provides a cloning or expression vector or plasmid comprising a nucleic acid sequence encoding the tumor antigen peptide of the disclosure. Alternatively, a nucleic acid encoding a tumor antigen peptide of the disclosure can be incorporated into the host cell genome. In either case, the host cell expresses the tumor antigen peptide or protein encoded by the nucleic acid. The term "host cell" as used herein refers not only to the particular subject cell, but to the progeny or potential progeny of said cell. A host cell can be any prokaryotic (eg, E. coli) or eukaryotic (eg, insect cells, cccznn / i znz / B / v yeast or mammalian cells) capable of expressing the tumor antigen peptides described in the present description. The vector or plasmid contains the necessary elements for the transcription and translation of the inserted coding sequence, and may contain other components such as resistance genes, cloning sites, etc. Methods that are well known to those skilled in the art can be used to construct expression vectors containing peptide or polypeptide-encoding sequences and suitable transcriptional and translational control / regulation elements operably linked thereto. These methods include in vitro recombinant DNA techniques, in vivo genetic recombination and synthesis techniques. Such techniques are described in Sambrook et al., (1989) Molecular Cloning, A Laboratory Manual, Cold Spring Harbor Press, New York) and Ausubel, F. M. et al., (1989) Current Protocols in Molecular Biology, John Wiley & Sons, New York. , N.Y. "Operably linked" refers to a juxtaposition of components, particularly nucleotide sequences, such that the normal function of the components can be performed. Thus, a coding sequence that is operably linked to regulatory sequences refers to a configuration of nucleotide sequences in which the coding sequences can be expressed under the regulatory control, ie, the transcriptional and / or translational control, of the regulatory sequences. Regulatory / control region or regulatory / control sequence, as used herein, refers to those non-coding nucleotide sequences that are involved in regulating the expression of a coding nucleic acid. Thus, the term "regulatory region" includes promoter sequences, regulatory protein binding sites, upstream activator sequences, and the like. In one embodiment, the nucleic acid (DNA, RNA) encoding the tumor antigen peptide of the disclosure is comprised or encapsulated within a vesicle, such as a liposome. In another aspect, the present disclosure provides an MHC class I molecule comprising (ie, presenting or binding to) a tumor antigen peptide. In one embodiment, the MHC class I molecule is an HLA-A2 molecule, in a further embodiment an HLA-A*02:01 molecule. In one embodiment, the MHC class I molecule is an HLA-A11 molecule, in a further embodiment an HLA-A* 11:01 molecule. In one embodiment, the MHC class I molecule is an HLA-A24 molecule, in a further embodiment an HLA-A*24:02 molecule. In another embodiment, the MHC class I molecule is an HLA-B07 molecule, in a further embodiment an HLA-B*07:02 molecule. In another embodiment, the MHC class I molecule is an HLA-B08 molecule, in a further embodiment an HLA-B*08:01 molecule. In another embodiment, the MHC class I molecule is an HLA-B40 molecule, in a further embodiment an HLA-B*40:01. In another embodiment, the MHC class I molecule is an HLA-C07 molecule, in a further embodiment an HLA-C*07:01 molecule. In one embodiment, the tumor antigen peptide is non-covalently bound to the MHC class I molecule (ie, the tumor antigen peptide is loaded or non-covalently bound to the peptide binding groove / pocket of the tumor molecule). MHC class I). In another embodiment, the tumor antigen peptide is covalently linked / linked to the MHC class I (alpha chain) molecule. In such a construct, the tumor antigen peptide and the MHC class I (alpha chain) molecule are produced as a synthetic fusion protein, typically with a short linker (eg, 5-20 residues, preferably around 8-12, eg , 10) flexible or spacer (for example, a polyglycine bond). In another aspect, the disclosure provides a nucleic acid encoding a fusion protein comprising a tumor antigen peptide defined herein fused to an MHC class I (alpha chain) molecule. In one embodiment, the MHC class I molecule (alpha chain)-peptide complex is multimerized. Accordingly, in another aspect, the present description provides a multimer of MHC class I molecule charged (covalently or not) with the tumor antigenic peptide mentioned in the present description. Said multimers can be linked to a label, eg a fluorescent label, which allows detection of the multimers. A large number of strategies have been developed for the production of MHC multimers, including MHC dimers, tetramers, pentamers, octamers, etc. (reviewed in Bakker and Schumacher, Current Opinion in Immunology 2005, 17:428433). MHC multimers are useful, for example, for the detection and purification of antigen-specific T cells. Therefore, in another aspect, the present description provides a method for detecting or purifying (isolating, enriching) CD8+ T lymphocytes specific for a tumor antigenic peptide defined in the present description, the method comprises contacting a cell population with a multimer MHC class I molecule charged (covalently or not) with the tumor antigenic peptide; and detecting or isolating CD8+ T cells bound by MHC class I multimers. CD8+ T cells bound by MHC class I multimers can be isolated using known methods, for example, fluorescence activated cell sorting (FACS) or magnetically activated cell sorting (MACS). In yet another aspect, the present disclosure provides a cell (eg, a host cell), in one embodiment an isolated cell, comprising the tumor antigen peptide, nucleic acid, vector or plasmid of the disclosure mentioned in the present disclosure, is that is, a nucleic acid or vector encoding one or more tumor antigen peptides. In another aspect, the present disclosure provides a cell expressing on its surface an MHC class I molecule (eg, an MHC class I molecule of one of the alleles described above) bound to or presenting a tumor antigen peptide according to the description. In one embodiment, the host cell is a eukaryotic cell, such as a mammalian cell, preferably a human cell, a cell line, or an immortalized cell. In another embodiment, the cell is an antigen presenting cell (APC) such as a dendritic cell (DC) or a monocyte / macrophage.In one embodiment, the host cell is a primary cell, a cell line, or an immortalized cell. Nucleic acids and vectors can be introduced into cells by conventional transformation or transfection techniques. The terms transformation and transfection refer to techniques for introducing foreign nucleic acid into a host cell, including calcium phosphate or calcium chloride coprecipitation, DEAE-dextran-mediated transfection, lipofection, electroporation, microinjection, and virus-mediated transfection. Suitable methods for transforming or transfecting host cells can be found, for example, in Sambrook et al., (suprd) and other laboratory manuals. Methods for introducing nucleic acids into mammalian cells in vivo are also known and can be used to deliver the vector or plasmid of the disclosure to a subject for gene therapy. Cells such as APCs can be loaded with one or more tumor antigen peptides using a variety of methods known in the art. As used herein, loading a cell with a tumor antigen peptide means that the RNA (mRNA) or DNA encoding the tumor antigen peptide, or tumor antigen peptide, is transfected into the cells or, alternatively, that the APC is transformed with a nucleic acid encoding the tumor antigen peptide. The cell can also be loaded by contacting the cell with foreign tumor antigen peptides that can bind directly to the MHC class I molecule present on the cell surface (eg, peptide-pulsed cells). Tumor antigen peptides can also be fused to a domain or motif that facilitates their presentation by MHC class I molecules, for example, to an endoplasmic reticulum (ER) retrieval signal, a C-terminal Lys-Asp-Glu-Leu sequence. (see Wang et al., Eur J Immunol. 2004 Dec;34(12):3582-94). In another aspect, the present disclosure provides a composition or combination / group of peptides comprising any one of, or any combination of, the tumor antigen peptides defined herein (or a nucleic acid encoding said peptide(s)). In one embodiment, the composition comprises any combination of the tumor antigen peptides defined herein (any combination of 2, 3, 4, 5, 6, 7, 8, 9, 10 or more tumor antigen peptides) or a combination of nucleic acids, encoding said tumor antigenic peptides. Compositions comprising any combination / subcombination of the tumor antigen peptides defined herein are encompassed by the present description. In one embodiment, the peptide composition or combination / set comprises at least 2, 3, 4, 5, 6, 7, 8, 9, 10 of the tumor antigen peptides comprising or consisting of the sequences set forth in SEQ IDs NO: 17-28. In one embodiment, the peptide composition or combination / set comprises at least 2, 3, 4, 5, 6, 7, 8, 9, 10 of the tumor antigen peptides comprising or consisting of the sequences set forth in SEQ IDs NOT: 29-39. In another embodiment, the combination or pool can comprise one or more known tumor antigens. Thus, in another aspect, the present disclosure provides a composition comprising any one of, or any combination of, the tumor antigen peptides defined herein and a cell expressing an MHC class I molecule (for example, an MHC molecule class I of one of the alleles described above). APCs for use herein are not limited to a particular cell type and include professional APCs such as dendritic cells (DCs), Langerhans cells, macrophages / monocytes, and B cells, which are known to present protein antigens in their cell surface, so as to be recognized by CD8+ T lymphocytes. For example, an APC can be obtained by inducing DC from peripheral blood monocytes and then contacting (stimulating) tumor antigen peptides, either in vitro, ex vivo, or in vivo. APC can also be activated to present a tumor antigen peptide. in vivo wherein one or more of the tumor antigen peptides of the disclosure are administered to a subject and APCs presenting a tumor antigen peptide are induced in the subject's body. The phrase "inducing an APC" or "stimulating an APC" includes contacting or loading a cell with one or more tumor antigen peptides, or nucleic acids encoding the tumor antigen peptides such that the tumor antigen peptides are presented on its surface by MHC class I. molecules. As indicated herein, according to the present disclosure, tumor antigen peptides can be loaded indirectly, for example, by using longer peptides / polypeptides comprising the sequence of tumor antigen peptides (including native protein ), which is then processed (eg, by proteases) within APCs to generate the tumor antigen peptide / MHC class I complexes on the surface of cells. After loading the APCs with tumor antigen peptides and allowing the APCs to present the tumor antigen peptides, the APCs can be administered to a subject as a vaccine. For example, in vivo administration may include the steps of: (a) harvesting APCs from a first subject, (b) contacting / loading the APCs from step (a) with a tumor antigen peptide to form MHC class complexes. 1 / tumor antigenic peptide on the surface of APCs; and (c) administering the peptide-loaded APCs to a second subject in need of treatment. The first subject and the second subject may be the same subject (eg, auto vaccine), or they may be different subjects (eg, allogeneic vaccine). Alternatively, in accordance with the present disclosure, there is provided the use of a tumor antigen peptide described in the present disclosure (or a combination thereof) for manufacturing a composition (eg, a pharmaceutical composition) for inducing antigen-presenting cells. Furthermore, the present disclosure provides a method or process for manufacturing a pharmaceutical composition for inducing antigen-presenting cells, wherein the method or process includes the step of mixing or formulating the tumor antigen peptide, or a combination thereof, with a pharmaceutically acceptable carrier. Cells such as APCs that express an MHC class I molecule (for example, an HLA-A2, HLA-A11, HLA-A24, HLAB07, HLA-B08, HLA-B40, or HLA-C07 molecule) loaded with any of, or any combination of, the tumor antigen peptides defined herein, can be used to stimulate / amplify CD8+ T cells, eg, autologous CD8+ T cells. Accordingly, in another aspect, the present disclosure provides a composition comprising any one of, or any combination of, the tumor antigen peptides defined herein (or a nucleic acid or vector encoding them); a cell expressing an MHC class I molecule and a T lymphocyte, more specifically a CD8+ T lymphocyte, for example, a population of cells comprising CD8+ T lymphocytes). In one embodiment, the composition further comprises a buffer, an excipient, a carrier, a diluent, and / or a medium (eg, a culture medium). In a further embodiment, the cccznn / i znz / B / v buffer, excipient, carrier, diluent and / or medium is / are pharmaceutically acceptable buffers, excipients, carriers, diluents and / or media (media). As used herein, pharmaceutically acceptable buffer, excipient, carrier, diluent, and / or medium includes any and all solvents, buffers, binders, lubricants, fillers, thickening agents, disintegrants, plasticizers, coatings, barrier coating formulations , lubricants, stabilizing agent, release retardant agents, dispersion media, coatings, antibacterial and antifungal agents, isotonic agents, and the like that are physiologically compatible, that do not interfere with the efficacy of the biological activity of the active ingredient(s). ) and are not toxic to the subject. The use of such media and agents for pharmaceutically active substances is well known in the art (Rowe et al., Handbook of Pharmacy excipients, 2003, 4th edition, Pharmaceutical Press, London, UK). Except insofar as any conventional media or agent is incompatible with the active compound (peptides, cells), their use in the compositions of the disclosure is contemplated. In one embodiment, the buffer, excipient, carrier, and / or medium is a non-naturally occurring buffer, excipient, carrier, and / or medium. In one embodiment, one or more of the tumor antigen peptides defined herein, or nucleic acids (eg, mRNA) encoding said one or more tumor antigen peptides, are comprised within or complexed with a liposome, eg, a cationic liposome (see, eg, Vítor MT et al., Recent Pat Drug Deliv Formul 2013 Aug;7(2):99-110). In another aspect, the present disclosure provides a composition comprising one or more of any, or any combination of, the tumor antigen peptides defined herein (or a nucleic acid encoding said peptide(s)), and a buffer, excipient, carrier, diluent, and / or medium. For compositions comprising cells (eg, APC, T-lymphocytes), the composition comprises a suitable medium that allows maintenance of viable cells. Representative examples of such media include saline, Earl's Balanced Salt Solution (Life Technologies®), or PlasmaLyte® (Baxter International®). In one embodiment, the composition (eg, pharmaceutical composition) is an immunogenic composition, vaccine composition, or vaccine. The term immunogenic composition, vaccine composition or vaccine as used herein refers to a composition or formulation that comprises one or more tumor antigenic peptides or vaccine vector and that is capable of inducing an immune response against one or more antigenic peptides. cccznn / i znz / B / v tumor cells present therein when administered to a subject. Vaccination methods for inducing an immune response in a mammal comprise the use of a vaccine or vaccine vector to be administered by any conventional route known in the field of vaccines e.g. mucosally e.g. ocular, intranasal, pulmonary, oral, gastric, intestinal, rectal, vaginal, or urinary tract route), parenteral route (for example, subcutaneous, intradermal, intramuscular, intravenous, or intraperitoneal) topical route (for example, via a transdermal delivery system such as a patch ). In one embodiment, the tumor antigen peptide (or a combination thereof) is conjugated to a carrier protein (vaccine conjugate) to increase the immunogenicity of the tumor antigen peptide(s). Therefore, the present disclosure provides a composition (conjugate) comprising a tumor antigen peptide (or a combination thereof) and a carrier protein. For example, the tumor antigen peptide(s) can be conjugated to a Toll-like receptor (TLR) ligand (see, eg, Zom et al., Adv Immunol. 2012, 114: 177- 201) or polymers / dendrimers (see, for example, Liu et al., Biomacromolecules. 2013 Aug 12;14(8):2798-806). In one embodiment, the immunogenic composition or vaccine further comprises an adjuvant. "Adjuvant" refers to a substance that, when added to an immunogenic agent such as an antigen (tumor and / or cell antigenic peptides according to the present disclosure), non-specifically enhances or potentiates an immune response to the agent in the host after exposure mix. Examples of adjuvants currently used in the vaccine field include (1) mineral salts (aluminum salts such as aluminum phosphate and aluminum hydroxide, calcium phosphate gels), squalene, (2) oil-based adjuvants such as emulsions of oil and surfactant-based formulations e.g. MF59 (oil-in-water emulsion stabilized with microfluidized detergent), QS21 (purified saponin), AS02 [SBAS2] (oil-in-water emulsion + MPL + QS-21), (3 ) particulate adjuvants, eg, virosomes (unilamellar liposomal vehicles incorporating influenza hemagglutinin), AS04 (aluminum salt [SBAS4] with MPL), ISCOMS (structured complex of saponins and lipids), polylactide co-glycolide (PLG), ( 4) microbial derivatives (natural and synthetic), for example, monophosphoryl lipid A (MPL), Detox (MPL + M. Phlei cell wall skeleton), AGP [RC529] (synthetic acylated monosaccharide), DC_Chol (lipoidal immunostimulators capable of self-organizing into liposomes), OM-174 (derived from lipid A), CpG motifs (synthetic oligonucleotides containing immunostimulatory CpG motifs), modified LT and CT (bacterial toxins engineered to provide non-toxic adjuvant effects), (5) immunomodulators endogenous humans, eg, hGM-CSF or hIL-12 (cytokines that can be delivered as encoded proteins or plasmids), Immudaptin (C3d tandem array) and / or (6) inert carriers, such as gold particles and the like. In one embodiment, the tumor antigen peptide(s) or composition comprising them is in lyophilized form. In another embodiment, the tumor antigen peptide(s) or composition comprising it is in a liquid composition. In a further embodiment, the tumor antigen peptide(s) is at a concentration of from about 0.01 pg / mL to about 100 pg / mL in the composition. In additional embodiments, the tumor antigen peptide(s) is at a concentration of from about 0.2 pg / mL to about 50 pg / mL, from about 0.5 pg / mL to about 10, 20, 30 , 40 or 50 pg / mL, from about 1 pg / mL to about 10 pg / mL, or about 2 pg / mL, in the composition. As indicated herein, cells such as APCs that express an MHC class I molecule loaded or bound to any of the tumor antigen peptides defined herein, or any combination thereof, can be used to stimulate / amplify T cells. CD8+ in vivo or ex vivo. Accordingly, in another aspect, the present disclosure provides T cell receptor (TCR) molecules capable of interacting with or binding to the complex of tumor antigen peptide / MHC class I molecule mentioned in the present disclosure, and nucleic acid molecules. encoding said TCR molecules, and vectors comprising such nucleic acid molecules. A TCR according to the present disclosure is capable of specifically interacting with or binding to a tumor antigen peptide carried or presented by an MHC class I molecule, preferably on the surface of a living cell in vitro or in vivo. A TCR and, in particular, nucleic acids encoding a TCR of the disclosure may, for example, be applied to genetically transform / modify T lymphocytes (eg CD8+ T lymphocytes) or other types of lymphocytes generating new T lymphocyte clones that specifically recognize a tumor antigen peptide / MHC class I complex. In a particular embodiment, T cells (eg, CD8+ T cells) obtained from a patient are transformed to express one or more TCRs that recognize a tumor antigen peptide and cells transformed cells are administered to the patient (autologous cell transfusion). In a particular embodiment, T cells (eg, CD8+ T cells) obtained from a cccznn / i znz / B / v donor are transformed to express one or more TCRs that recognize a tumor antigen peptide and the transformed cells are administered to a recipient (allogeneic cell transfusion). In another embodiment, the disclosure provides a T cell, eg, a CD8+ T cell transformed / transfected by a vector or plasmid encoding a tumor antigen peptide-specific TCR. In a further embodiment, the disclosure provides a method of treating a patient with autologous or allogeneic cells transformed with a tumor antigen peptide-specific TCR. In yet another embodiment, the use of a tumor antigen-specific TCR in the manufacture of autologous or allogeneic cells for the treatment of cancer is provided. In some embodiments, patients treated with the compositions (eg, pharmaceutical compositions) of the disclosure are treated before or after treatment with allogeneic stem cell transplantation (ASCL), allogeneic lymphocyte infusion, or autologous lymphocyte infusion. Compositions of the disclosure include: allogeneic T cells (eg, CD8+ T cell) activated ex vivo against a tumor antigen peptide; allogeneic or autologous APC vaccines loaded with a tumor antigen peptide; Tumor antigen peptide vaccines and allogeneic or autologous T lymphocytes (eg, CD8+ T lymphocyte) or lymphocytes transformed with a tumor antigen-specific TCR. The method of providing T cell clones capable of recognizing a tumor antigen peptide according to the disclosure can be generated and specifically targeted to tumor cells expressing the tumor antigen peptide in a subject (eg, graft recipient), eg, a ASCT and / or a donor lymphocyte infusion (DLI) recipient. Thus, the disclosure provides a CD8+ T cell encoding and expressing a T cell receptor capable of specifically recognizing or binding to a tumor antigen peptide / MHC class I molecule complex. ) can be a recombinant (engineered) or naturally selected T cell.Thus, this specification provides at least two methods of producing CD8+ T cells of the disclosure, comprising the step of contacting naive lymphocytes with a tumor antigen peptide / MHC class I molecule complex (typically expressed on the surface of cells). , such as APCs) under conditions conducive to triggering T-cell activation and expansion, which can be done in vitro or in vivo (i.e., in a patient administered an APC vaccine where the APC is loaded with an antigenic peptide tumor or in a patient treated with a tumor antigen peptide vaccine). By using a combination or group of tumor antigen peptides bound to MHC class I molecules, it is possible to generate a population of CD8+ T lymphocytes capable of recognizing a plurality of tumor antigen peptides. Alternatively, in vitro or ex vivo targeted or tumor antigen specific T cells can be produced / generated by cloning one or more nucleic acids (genes) encoding a TCR (more specifically alpha and beta chains) that specifically binds to a peptide complex. tumor antigen / MHC class I molecule (ie, recombinant or genetically modified CD8+ T cells). Nucleic acids encoding a tumor antigen peptide-specific TCR of the disclosure can be obtained using methods known in the art from a T cell activated against a tumor antigen peptide ex vivo (for example, with an APC loaded with a tumor antigen peptide); or from an individual exhibiting an immune response against the peptide / MHC molecule complex. The tumor antigen peptide-specific TCRs of the disclosure can be expressed recombinantly in a host cell and / or a host lymphocyte obtained from a graft recipient or graft donor, and optionally differentiated in vitro to yield cytotoxic T lymphocytes (CTLs). The nucleic acids (transgene(s)) encoding the TCR alpha and beta chains can be introduced into T cells (eg, from a subject to be treated or another individual) using any suitable method, such as transfection (by example, electroporation) or transduction (for example, through the use of a viral vector). Engineered CD8+ T cells expressing a TCR specific for a tumor antigen peptide can be expanded in vitro using well known culture methods. The present disclosure provides isolated CD8+ T cells that are specifically induced, activated, and / or amplified (expanded) by a tumor antigen peptide (i.e., a tumor antigen peptide bound to MHC class I molecules expressed on the cell surface), or a combination of tumor antigen peptides. The present disclosure also provides a composition comprising CD8+ T-lymphocytes capable of recognizing a tumor antigen peptide, or a combination thereof, according to the disclosure (ie, one or more tumor antigen peptides bound to MHC class I molecules). and said tumor antigen peptide(s). In another aspect, the present disclosure provides a cell population or cell culture (eg, a CD8+ T cell population) enriched for CD8+ T cells that specifically recognize one or more MHC class I molecule / tumor antigen peptide complexes as described in the present description. Such an enriched population can be obtained by performing an in vivo expansion of specific T lymphocytes using cells as APCs that express MHC class I molecules loaded with (eg, presenting) one or more of the tumor antigen peptides described herein. Enriched as used herein means that the proportion of tumor antigen-specific CD8+ T cells in the population is significantly higher relative to a native population of cells, i.e., that has not undergone an ex vivo expansion step of specific T lymphocytes. In a further embodiment, the proportion of tumor antigen peptide-specific CD8+ T cells in the cell population is at least about 0.5%, eg, at least about 1%, 1.5%, 2%, or 3%. In some embodiments, the proportion of tumor antigen peptide-specific CD8+ T cells in the cell population are about 0.5 to about 10%, about 0.5 to about 8%, about 0.5 to about 5%, about 0.5 to about 4%, about 0.5 to about 3%, about 1% to about 5%, about 1% to about 4%, about 1% to about 3%, about 2% to about 5%, about 2% to about 4%, about 2% to about 3%, about 3% to about 5%, or about 3% to about 4%. Such a cell population or culture (eg, a population of CD8+ T lymphocytes) enriched for CD8+ T lymphocytes that specifically recognize one or more MHC class I / peptide (tumor antigen peptide) complex molecules of interest can be used in cancer immunotherapy based on tumor antigens, as detailed below. In some embodiments, the tumor antigen peptide-specific CD8+ T cell population is further enriched, for example, through the use of affinity-based systems such as MHC class I molecule multimers loaded (covalently or not) with the peptide(s). ) tumor antigen(s) defined herein. Thus, the present disclosure provides a purified or isolated population of tumor antigen peptide-specific CD8+ T cells, eg, wherein the proportion of tumor antigen peptide-specific CD8+ T cells are at least about 50%, 60%, 70%, 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99%, or 100%. The present disclosure further relates to the use of any tumor antigen peptide, nucleic acid, expression vector, T-cell receptor, cell (eg, T-lymphocyte, APC) and / or composition according to the present disclosure, or any combination thereof, such as a drug or in the manufacture of a drug. In one embodiment, the medicament is for the treatment of cancer, eg, cancer vaccine. The present disclosure refers to any tumor antigen peptide, nucleic acid, expression vector, T cell receptor, cell (eg, T-lymphocyte, APC) and / or composition (eg, vaccine composition) according to the present description, or any combination thereof, for use in the treatment of cancer, eg, as a cancer vaccine (eg, therapeutic cancer vaccine).The tumor antigen peptide sequences identified herein can be used for the production of synthetic peptides suitable i) for in vitro priming and expansion of tumor antigen-specific T cells for injection into tumor patients and / or ii) as vaccines to induce or enhance the antitumor T cell response in cancer patients. In another aspect, the present disclosure provides the use of a tumor antigen peptide described herein, or a combination thereof (eg, a combination of peptides), as a vaccine to treat cancer in a subject. The present disclosure also provides the tumor antigen peptide described herein, or a combination thereof (eg, a combination of peptides), for use as a vaccine to treat cancer in a subject. In one embodiment, the subject is a recipient of tumor antigen peptide-specific CD8+ T cells. Accordingly, in another aspect, the present disclosure provides a method of treating cancer (eg, reducing the number of tumor cells, killing tumor cells), said method comprising administering (infusing) to a subject in need thereof an effective amount of CD8+ T cells that recognize (ie, express a TCR that binds) one or more MHC class I molecule / tumor antigen peptide complexes (expressed on a cell surface as an APC). In one embodiment, the method further comprises administering an effective amount of the tumor antigen peptide, or a combination thereof, and / or a cell (eg, an APC such as a dendritic cell) that expresses the molecule(s). ) of MHC class I loaded with the tumor antigenic peptide(s), to said subject after administration / infusion of said CD8+ T lymphocytes. In yet another embodiment, the method comprises administering to a subject in need thereof a therapeutically effective amount of a dendritic cell loaded with one or more tumor antigen peptides. In yet another embodiment, the method comprises administering to a patient in need thereof a therapeutically effective amount of an allogeneic or autologous cell expressing a recombinant TCR that binds to a tumor antigen peptide presented by an MHC class I molecule. In another aspect, the present disclosure provides the use of CD8+ T cells that recognize one or more MHC class I molecules loaded with (presenting) a tumor antigen peptide, or a combination thereof, to treat cancer (eg, reduce the number of tumor cells, kill tumor cells) in a subject. In another aspect, the present description provides the use of CD8+ T lymphocytes that recognize one or more MHC class I molecules loaded with (presenting) a tumor antigenic peptide, or a combination thereof, for the preparation / manufacturing of a medicament for treating cancer (eg, to reduce the number of tumor cells, kill tumor cells) in a subject. In another aspect, the present disclosure provides CD8+ T cells (cytotoxic T cells) that recognize one or more MHC class I molecules loaded with (presenting) a tumor antigen peptide, or a combination thereof, for use in the treatment of cancer. (eg, to reduce the number of tumor cells, kill tumor cells) in a subject. In a further embodiment, the use further comprises the use of an effective amount of a tumor antigen peptide (or a combination thereof) and / or a cell (eg, an APC) expressing one or more MHC class I molecules. loaded with (presenting) a tumor antigen peptide, after the use of said tumor antigen peptide-specific CD8+ T cells. The present disclosure also provides a method of generating an immune response against tumor cells expressing human MHC class I molecules loaded with any of the tumor antigenic peptides described herein or a combination thereof in a subject, the method comprises administering lymphocytes Cytotoxic T cells that specifically recognize MHC class I molecules loaded with the tumor antigen peptide or a combination of tumor antigen peptides. The present disclosure also provides the use of cytotoxic T lymphocytes that specifically recognize class I MHC molecules loaded with any of the tumor antigen peptides or combination of tumor antigen peptides described herein to generate an immune response against tumor cells expressing the MHC molecules. human class I cells loaded with the tumor antigen peptide or a combination thereof. In one embodiment, the methods or uses described herein further comprise determining the HLA class I alleles expressed by the patient prior to treatment / use, and administering or using tumor antigen peptides that bind to one or more of the alleles. of HLA class I expressed by the patient. For example, if a patient suffering from B-ALL is determined to express HLA-A2*01 and HLA-B*08:01, any combination of the tumor antigen peptides of (i) SEQ ID NO: 17-19, 27 and / or 28 (which binds to HLA-A2*01) and (ii) SEQ ID NO: 24 or 25 (which binds to HLA-B08*01) can be administered or used in the patient. In one embodiment, the tumor cells of the cancer to be treated, eg, leukemia or lung cancer, express one or more of the tumor antigen peptides described herein (SEQ ID NO: 17-39). In another embodiment, the methods or uses described herein further comprise determining whether the patient's tumor cells express one or more of the tumor antigen peptides described herein (SEQ ID NO: 17-39) and administering or using one or more of the tumor antigen peptide(s) expressed by the patient's tumor cells to treat the cancer. In one embodiment, the cancer is a blood or hematologic cancer, eg, leukemia, lymphoma, and myeloma. In one embodiment, the cancer is leukemia, including, but not limited to, acute lymphoblastic leukemia (ALL), acute myeloid leukemia (AML), chronic lymphocytic leukemia (CLL), chronic myeloid leukemia (CML), hairy cell leukemia (HCL) , T-cell lymphocytic leukemia (T-PLL), large granular lymphocytic leukemia, or adult T-cell leukemia. In another embodiment, the cancer is lymphoma including, but not limited to, Hodgkin lymphoma (HL), non-Hodgkin lymphoma (NHL), Burkitt lymphoma, precursor T-cell leukemia / lymphoma, follicular lymphoma, diffuse large B-cell lymphoma, mantle cell lymphoma, B-cell leukemia / chronic lymphocytic lymphoma, or MALT lymphoma. In a further embodiment, the cancer is a B-cell leukemia, such as B-ALL. In another embodiment, the cancer is a solid cancer, such as lung cancer. In a further embodiment, the lung cancer is non-small cell lung cancer (NSCLC). In one embodiment, the lung cancer is a squamous cell lung cancer (SQCLC), an adenocarcinoma, or a large cell anaplastic carcinoma (LCAC). In one embodiment, the tumor antigen peptides, nucleic acids, vectors, and compositions described herein can be used in combination with other therapies (eg, anti-tumor therapies), such as chemotherapy, immunotherapy (eg, CAR T cell-based therapy / NK, checkpoint inhibitor-based therapy, antibody-based therapy), radiotherapy, or surgery. Examples of immune checkpoint inhibitors include agents that inhibit PD-1, PD-L1, CTLA-4, KIR, CD40, TIM-3, or LAG-3, such as blocking antibodies. Examples of chemotherapy agents include, for example, amsacrine, bleomycin, busulfan, capecitabine, carboplatin, carmustine, chlorambucil, cisplatin, cladribine, clofarabine, chrysantaspase, cyclophosphamide, cytarabine, dacarbazine, dactinomycin, daunorubicin, docetaxel, doxorubicin, epirubicin, etoposide, fludarabine, fluorouracil (5-FU), gemcitabine, gliadel implants, hydroxycarbamide, idarubicin, ifosfamide, irinotecan, leucovorin, liposome doxorubicin, liposome daunorubicin, lomustine, melphalan, mercaptopurine, mesna, methotrexate, mitomycin, mitoxantrone, oxaliplatin, paclitaxel (Taxol), pemetrexed, pentostatin, procarbazine, raltitrexed, satraplatin, streptozocin, tegafur-uracil, temozolomide, teniposide, thiotepa, thioguanine, topotecan, treosulfan, vinblastine, vincristine, vindesine, vinorelbine, or a combination thereof. Alternatively, the chemotherapy agent may be a biologic agent, including Herceptin®(trastuzumab) against HER2 antigen, Avastin®(bevacizumab) against VEGF, or EGF receptor antibodies, such as Erbitux®(cetuximab) and Vectibix® ( panitumumab). Such additional agents or treatments may be administered / used before, during and / or after the administration / use of the tumor antigen peptides, nucleic acids, vectors and compositions described herein. Current treatments for ALL typically include vincristine, dexamethasone or prednisone, and an anthracycline drug such as doxorubicin (Adriamycin) or daunorubicin. Allogeneic stem cell transplantation (allo-SCT) is also performed in high-risk patients and patients with relapsed / refractory disease. Other agents in clinical development for the treatment of B-ALL include anti-CD22, anti-CD20, and anti-CD19 antibodies, as well as proteasome inhibitors (Bortezomib), inhibitors of the JAK / STAT signaling pathway (ruxolitinib), agent hypomethylating agent (Decitabine) and PI3K / mT0R inhibitors (see, for example, Terwilliger and Abdul-Hay, Blood Cancer J. 2017 Jun;7(6):e577). Current treatments for lung cancer typically include surgery, radiation therapy, chemotherapy with small molecular weight tyrosine kinase inhibitors (erlotinib, crizotinib), as well as immunotherapy with checkpoint inhibitors such as anti-PDI antibodies (pembrolizumab) ( see, for example, Dholaria et al., J Hematol Oncol. 2016;9:138). MQDQ(S) TO CARRY OUT THE INVENTION The present invention is illustrated in greater detail by the following non-limiting examples. Example 1: Materials and Methods Mice. C57BL / 6 mice were obtained from the Jackson Laboratory (Bar Harbor, ME). Mice were housed under specific pathogen-free conditions. cell lines. The EL4 T-lymphoblastic lymphoma cell line, the CT26 colorectal cancer cell line, and the HB-124 B-cell hybridoma were obtained from the American Type Culture Collection (ATCC). EL4 and CT26 cells were grown in RPMI 1640 / HEPES supplemented with 10% heat-inactivated fetal calf serum, 1% L-glutamine, and 1% penicillin-streptomycin. Cell culture media were further supplemented with 1% non-essential amino acids and 1% sodium pyruvate or 1% sodium pyruvate only for EL4 and CT26 cells, respectively. To produce the anti-CDR2 antibody, HB-124 cells were grown in IMDM supplemented with 10% heat-inactivated fetal bovine serum. Unless otherwise indicated, all reagents were purchased from Gibco®. Human primary samples. The primary leukemic samples (four B-ALL samples: 07H103, 10H080, 10H118, and 12H018) used in this study were collected and cryopreserved at the Banque de Cellules Leucémiques du Québec (BCLQ) at Hópital Maisonneuve-Rosemont. Primary leukemic samples were amplified in vivo after transplantation into NSG mice as previously described121. Briefly, 1-2 x 106 B-ALL cells were thawed and transplanted i.v. in 812-week-old sublethally irradiated (250 cGy, 137 Cs-gamma) NSG mice. Mice were sacrificed for signs of disease and cell suspensions were prepared from mechanically ruptured spleens or, for O7H1O3, from a mixture of splenocytes, bone marrow, and peritoneal ascites. Thereafter, Ficoll™ gradients were used to enrich for ALL-B cells prior to MAP isolation (see MAP Isolation section). Lung tumor biopsies (lc2, lc4 and lc6) were purchased from Tissue Solutions and homogenized prior to MAP isolation. (see the MAP Isolation section). For all samples, HLA typing was obtained using Optitype version 1.0, running with default settings for RNA sequencing (RNASeq) data (see RNA extraction, library preparation and sequencing section). cccznn / i znz / B / v Peptides. Native C13-labeled versions of TSA were synthesized by GenScript. Purity, as determined by the manufacturer, was greater than 95% and 75% for the native and C13-labeled peptides, respectively. Marine extraction of mTEChl. The thymus was isolated from 5-8 week old C57BL / 6 or Balb / c mice and mechanically disrupted to remove thymocytes. Stromal cell enrichment was performed as previously described2^ Thymic stromal cells were stained with biotinylated Ulex europaeus lectin 1 (UEA1; Vector Laboratories), PE-Cy7-conjugated streptavidin (BD Biosciences), and the following antibodies: Alexa Fluor™ 700 anti-CD45, PE anti-IAb (BD Biosciences), allophycocyanin-Cy7 anti-EpCAM (BioLegend). Cell viability was assessed using 7-aminoactinomycin D (7-AAD; BD Biosciences). Live mature mTEC (mTECh) was gated as 7-AAD' CD45' EpCAM+UEA1+MHC IIhl. The mTEChs were sorted on a three-laser ArialIIu FACS (BD Biosciences, Figure 13A). Human extraction of TEC and mTEC. Thymus was obtained from individuals 3 months to 7 years of age undergoing corrective cardiovascular surgery (CHU Saint Justine Research Ethics Board, protocol and biobank #2126). Briefly, the thymus was kept at 4 °C in 50 mL conical tubes containing medium and cut into 2-5 mm cubes hours after its surgical resection. For long-term preservation, thymus cubes were frozen in cryovials containing 10% heat-inactivated human serum / DMSO and maintained in liquid nitrogen for up to 3 years. Cryopreserved thymus samples were transferred to dry ice and used to isolate human TEC and mTEC following a protocol adapted from C. Stoeckle et al. Thymic tissue was cut into small pieces, then digested at 37 °C using a solution of 2 mg / mL Collagenase A (Roche) and 0.1 mg DNase I / mL (Sigma-Aldrich) in RPMI-1640. (Gibco) for three to five periods of 40 min. After the second digestion, a Trypsin / EDTA solution (Gibco) was added, for which the activity was neutralized by adding FBS (Invitrogen) 15 min before the end of the incubation. For TEC and mTEC sorting (Figure 13B), cell suspensions were stained with Pacific blue-conjugated anti-CD45 (BioLegend), PE-conjugated anti-HLA-DR (BioLegend), APC-conjugated anti-EpCAM (BioLegend ), anti-CDR2 conjugated with Alexa 488 (produced with the HB-124 hybridoma - see section Cell lines - and conjugated with Abcam's Dylight 488 Fast conjugation kit, for mTEC samples only) and cell viability was assessed using of 7-AAD (BD Biosciences). RNA extraction, library preparation and sequencing. For EL4 and CT26 cells, a replicate of 5 x 106 cells was used to perform RNA sequencing. For mTEC111de C57BL / 6 and Balb / c, RNA sequencing was performed in triplicate on a minimum of 31,686 or 16,338 FACS sorted cells taken from 2 females and 2 males. For primary leukemic cells, RNA-Seq was performed in a single replicate of 2.0 to 4.0 x 106 cells. For human TEC and mTEC, one replicate RNA-Seq was performed per donor with 33,076 to 84,198 FACS sorted TEC or 50,058 to 100,719 mTEC. In all cases, total RNA was isolated using TRIzol (Invitrogen), further purified using the RNeasy Kit or RNeasy Micro Kit (Qiagen) as recommended by each manufacturer. For each lung tumor biopsy (three in total), total RNA was isolated from -30 mg of tissues using the AllPrep DNA / RNA / miRNA Universal Kit (Qiagen) as recommended by the manufacturer and used to perform one replicate RNA-Seq per sample. Each murine sample (EL4, CT26, and murinahl mTEC) was quantified on a Nanodrop 2000 (Thermo Fisher Scientific) and RNA quality was assessed on a 2100 Bioanalyzer (Agilent Genomics) to select samples with an RNA integrity number > 9. For human samples (B-ALL, lung tumor biopsies, and human ECT / mTEC), total RNA quantification was performed using QuBit (ABI) and total RNA quality was assessed with the 2100 BioAnalyzer (Agilent Genomics) for select samples with RNA integrity number > 7. cDNA libraries were prepared from 24 pg for EL4 and CT26 cells, 50-100 ng for murine mTEChl, 500 ng for B-ALL samples, 4 pg for tumor biopsies lung, 8-13 ng for human TEC or 41-68 ng for human mTEC of total RNA using TruSeq Stranded Total RNA Library Prep Kit (EL4 cells), ΚΑΡΑ Stranded Kit mRNA-Seq (CT26 cells , C57BL / 6 mTECh, human mTEC, lung tumors, and B-ALL samples) or ΚΑΡΑ RNA HyperPrep Kit (Balb / c mTEChola, human TEC). These libraries were further amplified by 9-16 PCR cycles prior to sequencing. Paired-end RNA sequencing was performed on an Illumina NextSeq™ 500 (Balb / c mTECh, human TEC, and mTEC) or HiSeq™ 2000 (any other sample) and returned an average of 175 and 199 x 106 reads per murine and human sample , respectively. Generation of canonical cancer and normal proteomes. For all samples, RNA-Seq reads were trimmed for low quality 3' bases and sequencing adapters using Trimmomatic version 0.35 and then aligned to the reference genome, GRCm38.87 for murine samples and GRCh38.88. for human samples, using STAR version 2.5.1b4a running with default parameters except for the parameters -alignSJoverhangMin, —alignMatesGapMax, —alignlntronMax, and —alignSJstitchMismatchNmax for which the default values ​​were replaced with 10, 200,000, 200,000, and 5 - 1 5 5, respectively. Single base mutations with a minimum alternative count setting of 5 were identified using freeBayes version 1.0.2-16-gd466dde [arXiv:1207.3907] and exported in a VCF, which was converted to an SNP file format. pyGeno5a-compatible agnostic (available on GitHub, https: / / github.com / tariqdaouda / pyGeno. Finally, transcript expression was quantified in transcripts per million (tpm) with kallisto version 0.43.1 [Nicolas L Bray, Harold Pimentel , Páll Melsted, and Lior Pachter, Nearoptimal probabilistic RNA-seq quantification, Nature Biotechnology 34, 525-527 (2016)] running with default parameters.Notably, the kallisto index was constructed by using the index functionality and by using the appropriate *.cdna.all.fa.gz files downloaded from Ensembl To construct the canonical proteome of each sample, pyGeno was used to (i) insert high-quality sample-specific single base mutations (freeBayes quality > 20) in the reference genome, thus creating a custom exorn, and to (ii) export sample-specific sequence(s) of known proteins generated by expressed transcripts (tpm > 0). These protein sequences were written into a fasta file that was subsequently used for mass spectrometry (MS) database searches (canonical cancer proteome) and / or MHC I-associated peptide (MAP) sorting (canonical cancer proteome). normal canonical). See Figure IB for a schematic and Tables 4a-b for statistics. Generation of normal and cancer k-mer databases. For all normal and cancer samples, the fastq R1 and R2 files were independently downloaded and trimmed for low quality 3' bases and sequencing adapters using Trimmomatic version 0.35. To ensure that all reads were in the coding strand of the transcript, R1 reads were reverse complemented using the fastx_reverse_complement function of FASTX-Toolkit version 0.0.14. Using jellyfish 2.2.36a, k-mer databases of 33 and 24 nucleotides in length were generated and used for k-mer profiling and MAP sorting, respectively (see Figure 7A for details). Of note, when multiple biological replicates (murinahl mTEC) or when multiple unrelated donor samples (human TEC and mTEC) were available, the fastq files were concatenated to generate a single normal k-mer database per condition (C57BL / 6, Balb / c or human). K-mer filtering and generation of cancer-specific proteomes. To extract k-mers of 33 nucleotides in length that could give rise to TSA, the analysis was restricted to k-mers observed at least 4-fold in EL4 or CT26 k-mer cells, 7-fold in lung tumor biopsies, and 10 times in primary leukemic samples. Cancer-specific k-mers were then obtained by selecting those that were not expressed in the relevant mTEChl or human TEC / mTEC k-mer databases (see Figure 7B). This set of cancer-specific k-mers was assembled into longer linear sequences, called contigs. Briefly, one of the presented 33 nucleotide-long kmers is randomly selected to be used as a seed which is then spanned from both ends with consecutive overlapping k-mers by 32 nucleotides on the same strand (-r option disabled, as sets were used of k-mers chains). The assembly process stops when no k-mers can be assembled, i.e. no 32 nucleotide overlapping k-mers can be found, or when more than one k-mer fits (-an option of 1 for linear assembly) . If so, a new seed is selected and the assembly process resumes until all k-mers in the submitted list have been used once (Figure 7C). This step is performed by the kmer_assembly function of internally developed software called NEKTAR (https: / / bitbucket.org / eaudemard / nektar). To obtain proteins, a 3-frame translation of contigs that were at least 34 nucleotides in length was performed using a homebrew Python script. The cancer-specific proteins were then cleaved at internal stop codons and any resulting subsequences of at least 8 amino acids in length were given a unique identification before being included in the relevant database (see Figure 7D). See Figure IB for the schematic and Tables 4a-b for the statistics. MAP isolation. For EL4 and CT26 cells, three biological replicates of 250 x 106 exponentially growing cells were prepared from exponentially growing cells. For all primary leukemic samples, three biological replicates of ~450 to 700 x 106 cells were prepared from freshly collected leukemic cells (see Human Primary Samples section). MAPs were obtained as previously described7a, with minor modifications: after mild acid elution (MAE), peptides were desalted on an Oasis HLB cartridge (30 mg, Waters) and filtered at a 3 kDa molecular weight cutoff. (Amicon Ultra-4, Millipore) to remove β2-microglobulin (βιΜ) proteins. For one of the primary leukemic samples (specimen 10H080), four additional replicates of 100 x 106 cells were prepared and MAPs were isolated by immunoprecipitation (IP) as previously described18. Finally, lung tumor biopsies (wet weight between 771 and 1825 mg, see Human Primary Samples section) were cut into small pieces (cubes ~3 mm in size) and 5 mL of the cocktail was added to each tissue sample. protein inhibitor containing ice-cold PBS (Sigma). Tissues were first homogenized twice using an Ultra Turrax T25 homogenizer (20 seconds at 20,000 rpm, IKA-Labortechnik) and then once using an Ultra Turrax T8 homogenizer (20 seconds at 25,000 rpm, IKA -Labortechnik). Then, 550 µl of ice-cold 10X lysis buffer (10% w / v CHAPS) was added to each sample and MAPs were immunoprecipitated as previously described1 using 1 mg (1 mL) of W6 / 32 antibody covalently crosslinked to protein A magnetic beads per sample. Regardless of the MAP isolation technique, all MAP extracts were dried using a Speed-Vac and kept frozen prior to MS analyses. Mass spectrometry analysis. The MAP dry extracts were all resuspended in 0.2% formic acid. For EL4 and CT26, the MAP extracts were loaded onto a Cis housekeeping precolumn (5 mm x 360 μm i.d. packed with Jupiter Phenomenex Cis) and separated on a Cis housekeeping analytical column (15 cm x 150 pm i.d. packed with Jupiter Phenomenex Cis). with a 56 min gradient of 0-40% acetonitrile (0.2% formic acid) and a flow rate of 600 nL.min1 on an nEasy-LC II system. For all human samples, MAP extracts were loaded onto a casers C is analytical column (15 cm x 150 pm i.d. packed with Cis Jupiter Phenomenex) with a 56 min gradient of 0-40% acetonitrile (0.2 formic acid %, 07H103, 10H080-MAE, 10H118, and 12H018) or with a 100 min gradient of 5-28% acetonitrile (0.2% formic acid, lung tumor biopsies, and 10H080-IP) and a flow rate of 600 nl.min-1 on an nEasy-LC II system. Samples were analyzed with a Q-Exactive Plus (EL4, Thermo Fisher Scientific) or HF (all other samples, Thermo Fisher Scientific). For QExactive Plus, each full MS spectrum, acquired at 70,000 resolution, was followed by 12 MS / MS spectra, where the most abundant multi-charged ions were selected for 17 resolution MS / MS sequencing. 500, an autogain control target of le6, an injection time of 50 ms, and a collision energy of 25%. for the Q Exactive HF, each full MS spectrum, acquired at 60,000 resolution, was followed by 20 MS / MS spectra, where the most abundant multi-charged ions were selected for 15,000 resolution MS / MS sequencing (CT26, 07H103, 10H080-MAE, 10H118, 12H018) or 30,000 (lung tumor biopsies, 10H080-IP), an automatic gain control objective of 5 x 104, an injection time of 100 ms, and an energy of collision of 25%. Peptides were identified using Peaks 8.5 (Bioinformatics Solution Inc.) and peptide sequences were searched against the relevant global cancer database, obtained by concatenating the canonical cancer proteome and the cancer-specific proteome (see sections Generation of canonical cancer and normal proteomes and k-mer filtering and generation of cancer-specific proteomes). For peptide identification, tolerance was set at 10 ppm and 0.01 Da for the precursor ions and fragments, respectively. The occurrence of oxidation (M) and deamidation (NQ) were considered post-translational modifications. MAP identification. To select the MAPs, the unique ID lists obtained from Peaks were filtered to include peptides 8 to 11 amino acids in length that had a percentile rank < 2% as predicted by NetMHC 4.08a for at least one of the MHC I molecules. relevant. In addition, a local false discovery rate (FDR) of 5%, defined as the number of decoy identifications divided by the number of target identifications above a given Spike score threshold, was applied to limit the number of identifications of false positives in the final MAP, liza. Identification and validation of TSA candidates. To identify TSA candidates among all identified MAPs, an immunogenic status was assigned to each MAP / protein pair. To do so, each MAP and its associated MAP coding sequence(s) (MCS) were queried against the relevant normal or cancer proteome and normal or cancer k-mer databases of 24 nucleotides in length, respectively. MAPs detected in the normal canonical proteome were excluded regardless of their MCS detection status, as they are likely tolerogenic. MAPs that were truly cancer-specific, ie, not detected in the normal canonical proteome or normal k-mers, were marked as TSA candidates. MAPs absent from both canonical proteomes but present in both k-mer databases needed their MCS to be at least 10-fold overexpressed in cancer cells, relative to normal cells, in order to be marked as such (see cccznn). / i znz / B / v Figure 8A). Finally, MAPs encoded by several MCS (from different proteins) could only be marked as TSA candidates if their respective MCS were concordant, that is, if they consistently marked this MAP as a TSA candidate. MS / MS spectra of all TSA candidates were manually inspected to eliminate any false identification. In addition, sequences displaying multiple genomically possible I / L variants were further inspected to report both variants when distinguishable by MS, or only the most expressed variant when not (see Figure 8B). Finally, a genomic location was assigned to all those MS-validated TSA candidates by mapping MCS-containing reads to the reference genome (GRCm38.87 or GRCh38.88) by using BLAT (UCSC Genome Browser Tool). ). TSA candidates whose reads did not match a concordant genomic location or matched hypervariable regions (such as MHC, Ig, or TCR genes) or multiple genes were excluded. For those with a concordant genomic location, the Integrative Genome Viewer (IGV)9a was used to exclude TSA candidates with overlapping synonymous MCS mutations with respect to their relevant normal counterpart or, for human TSA candidates, those that overlap. to a known germline polymorphism (ie, listed in dbSNP v. 149, Figure 8C). The remaining peptides were classified as either mTSA or aeTSA candidates, depending on whether their MCS overlapped a cancer-specific mutation or not. Peripheral expression of MCS. To assess peripheral expression of MCS from TAA and aeTSA candidates, RNA-Seq data from (1) 22 murine tissues for which RNA10a'lla had been sequenced by the ENCODE system (Table 5) or (2) 28 tissues were used. peripheral humans (~50 donors per tissue), which had been sequenced by the GTEx system and downloaded from the GTEx Portal on 04 / 16 / 2018 (phs000424.v7.p2, Table 6). Briefly, RNA sequencing data from each tissue was transformed into 24 nucleotide-long k-mer databases with Jellyfish 2.2.3 (by using the -C option) and used to query the set of k -mer of 24 nucleotides from each MCS. For each RNA-Seq experiment, the number of reads that completely overlap a given MCS (faverian} was estimated by using the least occurrence of the k-mer set. Indeed, ~?averian was hypothesized because, except for low complexity RNA-Seq reads that might generate the same k-mer multiple times, a k-mer always originates from a single RNA-Seq read.Therefore, to compare the MCS expression level in all tissues, steroverlap value was transformed into number of reads detected per 108sequenced readsrP^m) by using the (roverlap X 10 ) rphm = ----------following formula :7 tot, conrfntque represents the total number of sequenced reads in a given RNA-Seq run. These values ​​were then log-transformed (^°Sio(rphm + 1)^ and averaged across all RNA-Seq experiments from a given tissue. aeTSA candidates exhibiting peripheral expression in 10 or fewer tissues (at rphm > 0) or in less than 5 tissues other than liver (at rphm > 15) for murine and human candidates, respectively, were considered to be genuine aeTSAs.The characteristics of those aeTSAs, as well as mTSAs, are reported in Tables la-b, 2a -d and 3a-c. MS validation of TSA candidates. For the CT26 TSA candidates and two EL4 TSA candidates (ATQQFQQL - SEQ ID NO: 11 and SSPRGSSTL - SEQ ID NO: 13), the previously acquired MS / MS spectra were compared with the corresponding 12C-analogs. For the other five EL4 TSA candidates evaluated in vivo (IILEFHSL - SEQ ID NO: 12, TVPLNHNTL - SEQ ID NO: 14, VNYIHRNV - SEQ ID NO: 15, VNYLHRNV - SEQ ID NO: 15, VTPVYQHL- SEQ ID NO: 16 ), MAP from six additional EL4 replicates (~450 to 1400 x 10 6 cells per replicate) were eluted and processed as described above (see Section MAP Isolation and Mass Spectrometry Analysis). For absolute quantification, three of six EL4 replicates were spiked with 500 fmol of each C-labeled 13TSA. For sequence validation, the MS / MS spectrum of 12C TSA candidates were acquired prior to sample analysis by PRM MS. . Briefly, the PRM acquisition, which monitored five peptides as scheduled (each peptide is only monitored in a 10-min window centered on its elution time), consisted of an MSI scan followed by targeted mode MS / MS scans. HCD. The automatic gain controls and injection times for the uplift scan and tandem mass spectra were 3e6 - 50 ms and 2e5 - 100 ms, respectively. In all cases, Skyline12ase was used to extract the endogenous MS / MS spectrum of each TSA candidate and compare it with the 12C MS / MS spectrum (sequence validation) or to extract the intensity of relevant endogenous and synthetic 13C-labeled peptides( absolute quantification). Using the following formula, these intensities were further used to calculate the TSA copy number per cell for each replicate: (nsynlhet.icx^endogenausx, / ) χ ( VjVn, \ .y.i..kiLj v ¿sjCQnt.syntaetw , initial number of moles and |· increased for the synthetic considered 13TSA labeled with C;and 'synthetic, intensity of the endogenous and TSA labeled with C, respectively;A, Avogadro's number;ceAS, initial number of cells used for elution mild acid. Cumulative number of transcripts detected in human TEC and mTEC samples. Restricting the analysis to transcripts expressed at a tpm > 1 in at least one of the six samples (2 TEC and 6 mTEC), Spearman's rank correlation coefficient was calculated for each 1 to 1 TEC / mTEC comparison. Then, using those same sets of expressed transcripts, the cumulative number of transcripts (c^) detected was calculated as each additional sample was analysed. Because the order in which samples are entered into the analysis can influence the values / 7^, values ​​across all sample permutations were averaged and those average data points were used to fit the following to X(nS - 1 ) cT = -----------+ c predictive curve (with the 'nls' function of the R):+“ -OI , with , the accumulated number of transcripts and the number of samples analyzed. This equation was then used to extrapolate the number of transcripts that would have been detected when studying up to 20 lim(cT) samples and that can be estimated simply by calculating Generation of bone marrow-derived dendritic cells (DC), mouse immunization and EL4 cell injection. Bone marrow-derived DCs were generated as previously described13a>14a. For immunization of mice, DC from male C57BL / 6 mice were pulsed with 2 pM of the selected peptide for 3 hours and then washed. They were injected i.v. to 8- to 12-week-old female C57BL / 6 mice with individually peptide-pulsed 106DC on days -14 and -7, or with irradiated (10,000 cGy) EL4 cells. As a negative control, female C57BL / 6 mice were immunized with non-pulsed DC. On day 0 and day 150, mice were injected i.v. with 5 x 105 EL4 cells and monitored for weight loss, paralysis, or tumor growth. IFN-γ ELISpot and avidity assays. ELISpot and avidity assays were performed as previously described142. Briefly, Millipore MultiScreen PVDF plates were permeabilized with 35% ethanol, washed and coated overnight using the ELISpot Ready-SET-Go! (eBioScience). On day 0 after immunization of the mice, splenocytes were harvested from immunized or naïve mice. 30 x 106 splenocytes / mL were stained with FITC-conjugated anti-CD8a (BD Biosciences) for 30 min at 4 °C, washed and sorted using a FACSAria™ IIu or FACSAria™ IIIu apparatus (BD Biosciences, Figure 13C ). Sorted CD8+ T cells were plated and incubated at 37 °C for 48 hours in the presence of irradiated (4000 cGy) splenocytes from syngeneic mice pulsed with the relevant peptide (4 pM for ELISpot assay and 10-4a 10- 14M for the avidity assay). As a negative control, CD8+ T cells from naïve mice were incubated with peptide-pulsed splenocytes. Spots were developed using the reagent set manufacturer's protocol and enumerated using an ImmunoSpot S5 UV Analyzer (Cellular Technology Ltd). IFN-γ production was expressed as the number of spot-forming units per 106 CD8+ T cells and EC50 was calculated using a dose-response curve. Lymphoid tissue cell isolation and tetramer-based enrichment protocol. Spleen and inguinal, axillary, brachial, cervical and mesenteric lymph nodes were collected from C57BL / 6 mice. Single cell suspensions were stained with Fe block and PE-labeled pMHC I tetramers or 10 nM APC (NIH Tetramer Core Facility) for 30 min at 4 °C. After washing with ice-cold sorting buffer (PBS with 2% FBS), cells were resuspended in 200 pL of sorting buffer and 50 pL of anti-PE and / or anti-APC antibody-conjugated magnetic microbeads (Miltenyi Biotech). , then they were incubated for 20 minutes at 4 °C. Cells were then washed and tetramer+ cells were magnetically enriched as previously described1521'16a. The resulting tetramer-spiked+fractions were stained with APC Fire 750-conjugated anti-B220, F4 / 80, CD19, CDllb, CDllc (BioLegend), PerCP-conjugated anti-CD4 (BioLegend), BV421-conjugated anti-CD3 (BD Biosciences), BB515-conjugated anti-CD8 antibodies (BD Biosciences), BV510-conjugated anti-CD44 antibodies (BD Biosciences), and Zombie NIR Repairable Viability Kit (BioLegend). Anti-CDllb and CDllc were left out for analysis of post-immunization repertoires because these markers can be expressed by some activated T cells17a'18a. The entire stained sample was then analyzed on a FACSCanto™ II cytometer (BD Biosciences) and fluorescent counting beads (Thermo Fisher Scientific) were used to normalize the results. As a negative control, repertoires of antigen-specific CD8+ T cells targeting 3 virus-derived antigens were enriched: gp-33 from lymphocytic choriomeningitis virus (LCMV) gp-33 protein (KAVYNFATC cccznn / i znz / B / v SEQ ID NO: 40; H-2Dsecund0), M45 from the murine cytomegalovirus M45 protein (HGIRNASFI SEQ ID NO: 41; H-2Db) and B8R from the vaccinia virus B8R protein (TSYKFESV-SEQ ID NO: 42; H-2Kb). Data. Information on all samples used in this study is listed in Table 7. The sequencing and expression data used in Figure 1 have been deposited in the NCBI Sequence Read Archive and GEO, which can be accessed from GEO with SuperSeries accession code GSE113992, containing accession code GSE111092 and GSE113972 for murine or human sequencing and expression data, respectively. The SuperSeries registry can be accessed through https: / / www.ncbi.nlm.nih.gov / geo / query / acc.cgi?acc=GSEl 13992, by entering the string cnutscacjbkzteb in the box. Raw data from MS and associated databases used in Figure 1 have been deposited with the ProteomeXchange Consortium via the PRIDE19a partner repository under the following dataset identifier: PXD009065 and 10.6019 / PXD009065 (CT26 cell line), PXD009064 and 10.6019 / PXD009064 (línea celular EL4), PXD009749 y 10.6019 / PXD009749 (07H103), PXD009753 y 10.6019 / PXD009753 (10H080, elución de ácido leve), ensayo- PXD007935 # 81756 y 10.6019 / PXD007935 (10H080, inmunoprecipitación)la, PXD009750 y 10.6019 / PXD009750 (10H118), PXD009751 and 10.6019 / PXD009751 (12h018), PXD009752 and 10.6019 / PXD009752 (LC2), PXD009754 and 10.6019 / PXD0075 (LC4) and PXD Example 2: Rationale and design of a proteogenomic method for the discovery of TSA. Attempts to computationally predict TSAs using various algorithms are plagued by extremely high false discovery rates27. Therefore, systems-level molecular definition of the MAP repertoire can only be achieved by high-throughput MS studies3. Current approaches use MS MS software tools, such as Peaks28, which rely on a user-defined protein database to match each acquired MS / MS spectrum to a peptide sequence. Since the reference proteome does not contain TSA, MS-based TSA discovery workflows must use proteogenomic strategies to create custom databases, derived from tumor RNA sequencing (RNA-Seq) data29, which ideally should contain all proteins, even those not annotated, expressed in the considered tumor sample. Since current MS / MS software tools cannot cope with the large search space created by translation of all frames from all RNA-Seq30,31 reads, a proteogenomics strategy was devised that enriches for RNA-Seq-specific sequences. cancer in order to comprehensively characterize the landscape of TSAs encoded by all genomic regions. The resulting database, called the Global Cancer Database, is made up of two customizable parts. The first part, called the cancer canonical proteome (Figure 1A), was obtained by in silico translation of transcripts encoding proteins expressed in their canonical framework; therefore, it contains proteins encoded by exonic sequences that are normal or contain single base mutations. The second part, called the cancer-specific proteome (Figure IB), was generated using an unaligned RNA-Seq workflow, called the k-mer profile, because current mappers and variant identifiers misidentify variants. structural. This second data set allowed detection of peptides encoded by any reading frame from any genomic origin (including structural variants), as long as they were cancer-specific (ie, absent from normal cells). Here, we chose to use mTEChi as a normal control because they express most of the known genes and induce central tolerance to the MAPs encoded by their vast transcriptome32. Therefore, to identify RNA sequences that were cancer-specific, cancer RNA-Seq reads were split into sequences 33 nucleotides in length, called k-mers33, from which syngeneic mTEChi k-mers were removed (Las Figures 7A-B). The redundancy inherent in the k-mer space was removed by assembling overlapping cancer-specific k-mers into longer sequences, called contigs, which were further 3 frames translated in silico (Figure IB and Figures 7C-D). The canonical cancer proteome and the cancer-specific proteome were concatenated to create a global cancer database, one for each sample tested. Using such optimized databases, MAPs eluted from two well-characterized mouse immoral cell lines, specifically CT26, a colorectal carcinoma from a Balb / c mouse and EL4, a T-lymphoblastic lymphoma from a C57BL / 6 mouse, were identified that were sequenced by MS. (Figure 1C). Example 3: Non-coding regions are the main source of TSA. At a false discovery rate of 5%, 1,875 MAPs were identified in CT26 cells and 783 MAPs in EL4 cells. Among them, MAP proteomes absent from the mTECh were considered candidates for TSA if (i) their 33 nucleotide-long MAP coding sequence (MCS), derived from a full length cancer-restricted 33 nucleotide k-mers, was absent from the mTECh. mTEChi transcriptome or si ) its 24- to 30-nucleotide-long MCS, derived from a cccznn / i znz / B / v truncated version of a cancer-restricted 33-nucleotide-long k-mers, was overexpressed at least 10-fold in the cancer transcriptome vs. mTEChl cells (Figure 8A). Following the MS-related validation steps and genomic location assignment (Figure 8B-C), a total of 6 mTSA and 15 aeTSA candidates were obtained: 14 presented by CT26 cells and 7 by EL4 cells (Figure 2A). -B). MAPs that were mutated and aberrantly expressed were included in the mTSA category. All of these MAPs are believed to be new and absent from the immune epitope database36, except for one: the AHI peptide (SPSYVYHQF), the only 150 aeTSA previously identified in CT26 cells by reverse immunology9,37. To assess the rigor of the database creation strategy based on the deletion of kmers from mTEChde cancer k-mers, peripheral expression of the MCS encoding aeTSA was assessed in a panel of 22 tissues38,39 (Table 5). Four of the 15 aeTSA candidates had an expression profile similar to previously reported overexpressed tumor-associated antigens (TAAs),40,41 as their MCS were expressed in most or all tissues (Figure 2C). Therefore, these four peptides were excluded from the TSA list. In contrast, 11 MAPs were considered to be genuine aeTSAs since their MCSs were completely absent or present in trace amounts in some tissues (Figure 2C). In fact, detection of low levels of transcripts is negligible since MAPs are preferentially derived from highly abundant transcripts.42,43. This concept is illustrated by AHI TSA, which elicits strong antitumor responses without adverse effects9,37, despite weak expression of its MCS in the liver, thymus, and urinary bladder (Figure 2C). These results demonstrate that subtraction of mRNA sequences found in mTEChlen strongly enriches for cancer-restricted MCS. When considering the entire murine TSA dataset (6 mTSA and 11 aeTSA), the most prominent finding is that most of them derive from atypical translation events: out-of-frame translation of the coding exon or translation of non-coding regions. (Figure 2D). Furthermore, all but two of the identified TSAs would have been lost with classical exorn-based approaches, since their source sequence is not annotated as protein-coding. Interestingly, it was also observed that any type of non-coding region can generate TSA (Table 1): intergenic and intronic sequences, non-coding exons, UTR / exon junctions, as well as ERE, which appear to be a particularly rich source of TSA (8). aeTSA and 1 mTSA). Finally, the approach described in the present description efficiently captured structural variants as an antigen, it was identified VTPVYQHL, derived from a very large (~7500 bp) intergenic deletion in EL4 cells (Table Ib). Taken together, these observations confirm that noncoding regions are the major source of TSA and that they have the potential to considerably expand the TSA tumor landscape. Further studies were performed on some of the TSAs that seemed most attractive, ie those that are presented by EL4 cells and whose MCS is not expressed by any normal tissue (Figure 2C and Table Ib). To assess their immunogenicity, C57BL / 6 mice were immunized twice with non-pulsed (control group) or TSA-pulsed DC before challenge with live EL4 cells. Priming against IILEFHSL or TVPLNHNTL prolonged the survival of 10% of the mice, and only the mouse immunized with TVPLNHNTL survived to day 150 (Figure 3A). The other three TSAs showed survival rates to day 150 of 20% (VNYIHRNV), 30% (VTPVYQHL), and 100% (VNYLHRNV) (Figure 3B, C). To assess the long-term efficacy of TSA vaccination, surviving mice were boosted with live EL4 cells at day 150 and monitored for signs of disease. The two survivors immunized with VNYIHRNV died of leukemia within 50 days, while all the others (immunized against TVPLNHNTL, VTPVYQHL or VNYLHRNV) survived challenge (Figure 3). Therefore, it can be concluded that immunization against individual TSAs confers different degrees of protection against EL4 cells and that, in most cases, this protection is long-lasting. Example 4: Frequency of TSA-specific T cells in naïve and immunized mice. In several models, the resistance of the immune response in vivo is regulated by the number of T cells reactive to the antigen44,45. Therefore, the frequency of TSA-specific T cells in naïve and immunized mice was assessed using a tetramer-based enrichment protocol46,47, for which the gating strategy and a representative experiment can be found in FIG 9A. -C. As positive controls, highly abundant CD8 T cells specific for three virus epitopes (gp-33, M45 and B8R) were used, and their frequencies were confirmed to be within the range of those observed in previous studies45 (Figure 4A). In naïve mice, CD8 T cells specific for TVPLNHNTL, VTPVYQHL, and IILEFHSL were rare (less than one tetramer + cell per 106 CD8 T cells), whereas CD8 T cells specific for ERE TSAs (VNYIHRNV and VNYLHRNV) showed frequencies similar to those of viral controls (Figure 4A and Figure 10A). Consequently, in mice immunized with TSA-pulsed DC, the frequencies of T cells against the two ERE TSAs were found, as assessed by tetramer staining or IFN-γ ELISpot assays (FIG 9C-D and 10A). ), were significantly higher than those of TVPLNHNTL, VTPVYQHL and IILEFHSL (Figure 4B-C). Furthermore, in both naïve and immunized mice, antigen-specific T cell frequencies were found to be highly correlated (FIG 11A-C). Finally, the functional avidity of T cells specific for VNYIHRNV and VNYLHRNV was estimated to be similar to that of T cells specific for two highly immunogenic non-self antigens: the minor histocompatibility antigens H7a and H13a (Figure 4D). Therefore, these TSAs, derived from putative non-coding regions, were recognized by highly abundant T cells with high functional avidity. This is particularly noteworthy for VNYIHRNV aeTSA as it has an unmutated germline sequence. Taken together, these results show that the frequency of TSA-specific T cells is generally a significant parameter for TSA immunogenicity. However, VTPVYQHL provided the second best protection against EL4 challenge despite the fact that its cognate T cells were present at a very low frequency (FIG 3 and 4A-C). To better assess the importance of T cell expansion in protection against leukemia, the frequency of tetramer+CD8 T cell presence was estimated in long-term survivors following rechallenge with EL4 cells at day 150 (Figure 3). ). These analyzes were performed on day 210 or at sacrifice (in the case of VNYIHRNV-primed mice). All long-term survivors, including mice immunized with VTPVYQHL, showed a conspicuous population of (tetramer4-) CD8 T cells (FIG 10B-C). Although VNYIHRNV was recognized by a large tetramer + cell population, this was not sufficient to protect mice upon rechallenge under the experimental conditions used in the present description. Example 5: The importance of antigen expression for protection against EL4 cells. Next, the impact of antigen expression on immunogenicity was assessed by assessing the abundance of TSA at the RNA level in the EL4 cell population that was injected on day 0 (Figure 3). The sequence encoding the TSA conferring the best protection against EL4 cells (VNYLHRNV) was found to be expressed at a much higher level than the other TSAs (Figure 5A). This suggests that VNYLHRNV is probably clonal (expressed by all EL4 cells) and highly expressed, whereas the other TSAs are subclonal and / or expressed at low levels. Next, using parallel reaction monitoring (PRM) MS, the number of TSA copies per cell in the population of EL4 cells used for rechallenge (day 150, Figure 5B) was analyzed. There was no linear relationship between the abundance of TSA in the RNA and the level of peptide40 (Figure 5A-B). In particular, the best TSA, VNYLHRNV, was one of the two most abundant TSAs (>500 copies per cell), while VNYIHRNV, which did not offer significant protection upon rechallenge under the experimental conditions used in the present description (Figure 3B), was no longer detected in EL4 cells. This observation suggests that VNYIHRNV was a subclonal TSA and that loss of antigen most likely explained the lack of significant protection upon rechallenge. Finally, it was observed that TSAs were immunogenic when presented by DCs, but not when presented by EL4 cells: i) injection of live EL4 cells without prior immunization did not induce a significant expansion of TSA-specific T cells, and ii ) immunization with irradiated EL4 cells did not confer significant protection against live EL4 cells (Figure 5C-D and Figure 10D). This suggests that, in the absence of immunization, highly immunogenic TSAs (such as VNYLHRNV) were ignored because they were not efficiently cross-presented by DCs, highlighting the importance of effective T-cell priming in immunotherapy. of cancer. Example 6: Non-coding regions expand the TSA landscape of human primary tumors. Having established that non-coding regions are the major source of TSA in two murine cell lines, the proteogenomic approach described in the present disclosure was applied to seven human primary tumor samples: four B-lineage ALL and three lung cancers. To do so, instead of using RNA-Seq data from murine syngeneic mTEC111, we sequenced the transcriptome of total TECs (n = 2) and purified mTECs (n = 4) from six unrelated donors undergoing corrective cardiovascular surgery. In particular, minimal differences between individuals were found, and this cohort size was shown to be sufficient to cover almost the entire mTEC transcriptomic landscape (FIG 12A-B). By using these RNA-Seq data as the repertoire of normal k-mers for the workflow described in Figure 1, 3 mTSA and 27 aeTSA candidates were identified (Figure 6A). In addition to being extensively validated, mTSAs were also guaranteed not to interbreed with known germline polymorphisms. To further validate the status of aeTSA candidates, the expression of aeTSA MCS in RNA-Seq data from 28 tissues (6-50 individuals per tissue, Figure 6B and Table 6), similar to what was done for murine aeTSAs (Figure 2C). Based on these data, six aeTSA candidates were excluded: i) three were widely expressed, like the majority of 5 previously reported overexpressed TAAs48, and ii) three were expressed at significant levels in only one organ, the liver (Figure 1). 6B). Thus, a total of three mTSA and 20 non-redundant aeTSA candidates were identified (Figure 6C and Tables 2a-d and 3a-c). Of note, the SLTALVFHV aeTSA was shared by the two HLA-A*02:01-positive ALLs (Tables 2a and 2d). This aeTSA is derived from the 3'UTR of TCL1A, a gene implicated in lymphoid neoplasms. Taken together, the results show that the proteogenomic approach described in the present disclosure can characterize the mTSA and aeTSA repertoire in individual tumors in approximately two weeks. BOARDS Table the; CT26 TSAs TSA Sequence (SEQ ID NO) Genomic Position Ensembl Transcript ID SA T Origin Frame Molécuk MHCI i Percentile Rank GYQKMKALL (SEQ ID NO: 1) chr8: 123429315- 123429341 MuLV ERE H-2-Kd 0.06 KPLK / EAPLDL (SEQ ID NO:2) chrl: 173783238- 173783264 ENSMUST00000155076 Intron H-2-Ld 0.01 KYLSVQS / GQL (SEQ ID NO:3) chrl7:29332770- 29332778 | chrl7:29333514- 29333531 ENSMUST00000095427 Coding exon In-frame H-2-Kd 0.01 KYLSVQS / GQLF (SEQ ID NO:4) chrl7:29332767- 29332778 | ENSMUST00000095427 Encoding exon H-2-Kd 0.25 chrl7:29333514cccznn / i znz / B / v In the frame 29333531 LPQELPGLVVL (SEQ ID N0:5) chr8: 123427101- MuLV 123427133 ERE H-2-Ld 0.5 MPHSLLPLVTF (SEQ ID N0:6) chr7: 89664573- ENSMUST00000159167 89664605 Intron H-2-Ld of QG2 Exon 0.PMRI LF (SEQ ID N0:7) chr9: 66126885- ENSMUST00000034945 66126911 encoding Outside H-2-Dd 0.05 frame SGPPYYKGI (SEQ ID N0:8) chr8: 121839803- MMERGLN_I or 121839829 ENSMUST000000RE2 Intrond H-D6 0.25 SPHQVFNL (SEQ ID NO:9) chr8: 123428239- MuLV 123428262 ERE H-2-Ld 0.01 SPSYVYHQF (SEQ ID NO: 10) chr8: 123426985- MuLV 123427011 ERE H-2-Ld 0.5 Table IB: TSAs EL4 TSA Sequence (SEQ ID NO) Genomic Position Transcript ID Molecule Frame MHCI Percentile Rank of Ensembl origin TSA ATQQFQQL (SEQ ID chr8: 123426867- MuLV ERE H-2-Kb 0.2 NO:11) IILEFHSL (SEQ ID NO: 12) 123426844 chrlO: 116678525- 116678548 ENSMUST000001816 56 non-coding exon H-2-Kb 0.02 SSPRGSSTL chr6: B3A or (SEQ ID 114732754- ENSMUST000000324 ERE or Intron H-2-Db 0.3 NO:13) 114732780 57 TVPLNHNTL (SEQ ID NO: 14) chr4: 83615597- 83615624 ENSMUST00000053 Novel2-D antisense H-4b 0.12 VNYI / LHRNV chr4: (SEQ ID 46583174- MMTV ERE H-2-Kb 0.01 NO:15) 46583197 VNYI / LHRNV chr4: (SEQ ID 46583174- MMTV ERE H-2-Kb 0.01 NO: 15) 46583197 VTPVYQ[HL (SEQ ID NO: 16) chr2:75078751- 75078756 [ chr2:75086270- 75086287 N / A Intergenic H-2-Kb 0.01 Table 2a: Characteristics of B-ALL specimens detected with human TSAs - TSAs 07H103 TSA Sequence (SEQ ID NO) Transcript ID Position Molecule Frame Percentile Rank Genomics of Ensembl origin T 'SA MHCI KILILLQSL (SEQ ID NO: 17) chr5: 132450600- L1ME3G or 132450626 ENST00000407797 ERE Intron 0 A*02:01 1, 8 KISLYLPAL (SEQ ID NO:18) chr8: 144861684- LTR46-int 144861710 ERE A*02:01 0.5 SLTALVFHV (SEQ ID NO: 19) chrl4: 95710533- ENST00000554012 95710559 3'UTR A*02:02 06 Table 2b: Characteristics of B-ALL specimens detected with human TSA - TSA cccznn / i ζπζ / β / υιλι 10H080 TSA Sequence (SEQ ID NO) Ensembl Genomic Transcript ID Position MHCI Molecule TSA Origin Frame Percentile Rank HETLRLLL (SEQ ID NO:20) chr6: 106197722- ENST00000369076 106197745 Intron B*40:01 1.2 RIFGFRLWK (SEQ ID NO:21) chrl: 80641339- ENST00000418041 80641365 Intron A*ll:01 0.01 TSFAETWMK (SEQ ID NO:22) chr7: 43947484- L1PA6 or 43947510 ENST00000427076 Intron ERE or A*llQLTSIPK0,ll0:1 NO:23) chr2: 237428272- N / A 237428298 Intergenic A* 11:01 0.15 Table 2c: Characteristics of B-ALL Specimens Detected with Human TSA - TSA 10H118 Sequence ID TSA Molecule Frame Position Rank (SEQ ID NO) transcript of genomic origin TSA MHC I percentile Ensembl LPFEQKSL chr2: 47522843- (SEQ ID ENST00000327876 Intron B*08:01 0.7 47522866 NO:24) SLREKGFSI chrl: ENST00000367667 Intron B *08:01 0.09 (SEQ ID 175955400- ΝΟ:25) 175955426 VPAALRSL chr7: (SEQ ID 106886341- ENST00000359195 Intron B*07:02 0.3 NO:26) 106886364 Table 2d: Characteristics of B-ALL Specimens Detected with Human TSA - TSA 12H018 ID's TSA Sequence Molecule Frame Position Transcript Rank of (SEQ ID NO) genomic origin TSA MHC I percentile ensemble LLAATILLSV chr2: (SEQ ID NO:27) SLFVA / VSLSL 174631426- ENST00000392547 Intron A*02:01 0.2 174631455 chr6: Exon of (SEQ ID NO:28) SLTALVFHV 106971679- ENST00000606017 coding A*0.6 106971705 In the chrl4 framework: (SEQ ID NO: 19) 95710533- ENST00000402399 3'UTR A*02:01 0.06 95710559 Table 3a: Characteristics of human TSAs detected in lung tumor biopsies lc2 TSAs TSA Sequence (SEQ ID NO) Genomic Position Transcript ID Ensembl of Frame of Origin TSA Molecule MHCI Percentile Rank IIAPPPPPK (SEQ ID NO:29) chrl4: 21098919- 21098945 ENST00000421093 5'UTR A*ll:01 0.15 LVFNIILHR chr6: 6800963- N / A Intergenic A*ll:01 0.25 (SEQ ID 6800989 NO:30) MISPVLALK (SEQ ID NO:31) chrl9: 41751004- ENST00000595740 41751030 5'UTR A*ll:01 0.03 SLSYLILKK Exon of (SEQ ID NO:32) chrX: 107212979- EN3050103 coding Outside A*ll:01 0.05 SSASQLPSK framework (SEQ ID NO:33) chrló: 19430493- L4_B_Mam or 19430519 ENST00000542583 ERE 5'UTR or A*ll:01 0.07 SVIQTGHLAK (SEQ ID NO:34) chr3: 0.03 Table 3b: Characteristics of Human TSAs Detected in lc4 TSA Lung Tumor Biopsies TSA Sequence (SEQ ID NO) Genomic Position Transcript ID Ensembl of Molecule MHCI Percentile Rank of TSA Origin Framework KPSVFPLSL chrl4: (SEQ ID 37589683- N / A Intergenic B *07:02 0.15 NO:36) 37589708 Table 3c: Characteristics of human TSAs detected in lung tumor biopsies lc6 TSAs TSA Sequence (SEQ ID NO) Genomic Position Transcript ID Ensembl of Frame of Origin TSA Molecule MHCI Percentile Rank QR / KF / LQGRVTM (SEQ ID NO:37) chrl5: 19972868- 19972894 N / A Intergenic C*07:01 0, 02 SRFSGVPDRF (SEQ ID NO:38) chr2: 89234284- 89234313 N / A Intergenic A*24:02 0.9 TYTQN / DFNKF (SEQ ID NO:39) chrll: 14968916- 14968942 ENST00000331587 Coding exon In A frame* 24:02 0.03 Table 4a; Statistics related to the generation of the world cancer databases - murine samples EL4 mTEChi_C57BL / 6 CT26 mTEChi_Balb / c Expresado 64318 86947 65242 82420 (tpm> 0) Proteomas transcriptos Codificación canónicos 34171 47086 35104 44943 de proteínas proteínas Distinto 35280 50304 37810 54456 lecturas Total 240372644 456991966 247522370 455625158 Total 14862978110 28980506746 15026027458 21482018335 Distinto 429163639 1084732266 507092097 1115569754 k-mers Cancer Count > 4 116852296 104699335 Proteomes (k = 33nts) Specific 19091379 22892864 cancer ____________ ..... Distinct 895313 1845144 contigs > 34nts 715161 1377631 Distinct proteins 7a709, > 6 2151709 Table 4b: Statistics related to the generation of global cancer databases - human samples 07H103 10H080 10H118 EXPRESSED (TPM> 107590 115494 116981 0) PROTEOMS TRANSCRIPTS CODING OF PARTIAL PROTEINS 57931 62280 63133 DIFFERENT PROTEINS 59082 64150 63921 READINGS TOTAL 105 863 640 129 444 492 Distinct 633 011 468 761 444 095 1 119 514 550 k-mers (k = Proteomes 33nts) Count > 7 or 10 77 745 744 98 652 247 135 682 880 specific Cancer specific Π 694 475 20 232 820 4070 Different 492 273 778 594 1 412 680 > 34nts 440 367 697 184 1 246 048 proteins Different, > 8aa 1 326 854 2 156 187 3 708 759 12H018 LC2 LC4 expressed (TPM> Transcripts 113438 116600 117476 0) Proteomas Coding of canonical proteins 61481 66874 67549 Different proteins 63767 70493 71734 Readings Total 161 724 658 268 396 930 262 531 548 17 197 19 197 19 197 19 197 177 specific k-mers (k = Distinct 868 719 740 669 751 679 727 571 721 from cancer 33nts) Count > 7 or 10 96 193 003 78 611 668 81 410 185 17 879 385 9 003 818 78 918 specific cancer contigs Distinct > 34nts 758 491 666 164 669 145 513 928 749 712 581 510 proteins Distinct, > 8aa 2 014 334 1 401 735 1 554 082 LC6 102015 062015 Transcribed Canonical Proteomas Expressed (TPM> 0) Protein coding 119870 67135 62976 37073 85686 49155 Different proteins 71526 46181 67497 Readings Total 246 868 078 134 624 214 136 558 238 Total 16 284 413 566 (k = 33nts) Distinct Count > 7 or 10 Cancer specific 864 050 270 97 121 823 17 663 050 Distinct contigs > 34nts 1 113 278 886 470 proteins Distinct, > 8aa 2 431 066 S5 S9 S10 SIL expressed 0) (TPM> 95090 118246 112739 119225 PROTEOMS TRANSCRIPTS CODING OF CANONICAL PROTEIN 55103 66223 63113 66695 DIFFERENT PROTEINS 70276 79469 75384 8099 PROTEOMS TOTAL READINGS TOTAL 2003 Different b b b b Count > 7 or 10 Cancer-specific contigs Distinct > 34nts Distinct proteins, > 8aa Table 5: Accession numbers of the ENCOPE data sets used in this study Tejido Números de acceso (SRA) Tejido adiposo SRR5171088, SRR5171089 Glándula suprarrenal SRR5171111, SRR5171112, SRR5047957, SRR5047958, SRR5047959, SRR5047960, SRR5047961, SRR5047962 Cerebro SRR5171101,SRR5171102 Colon SRR5047913, SRR5047914, SRR5047915, SRR5047916, SRR5047917, SRR5047918 Duodeno SRR5047963, SRR5047964, SRR5047965, SRR5047966, SRR5047967, SRR5047968, SRR5047969 Almohadilla de grasa gonadal SRR5047970, SRR5047971, SRR5047972, SRR5047973 Corazón SRR5171076, SRR5171077, SRR5047921, SRR5047922, SRR5047923, SRR5047924 Riñón SRR5047925, SRR5047926, SRR5047927, SRR5047928, SRR5047929, SRR5047930, SRR5171094, SRR5171095 Intestino grueso SRR5047975, SRR5047976, SRR5047977, SRR5047978 Hígado SRR3192469, SRR3192470, SRR5171078, SRR5171079, SRR5047931, SRR5047932, SRR5047933, SRR5047934, SRR5047935, SRR5047936 Pulmón SRR5171113, SRR5171114, SRR5047937, SRR5047938, SRR5047939, SRR5047940 Glándula mamaria SRR5047979, SRR5047980, SRR5047981, SRR5047982, SRR5047983, SRR5047984 Ovario SRR5047985, SRR5047986, SRR5047987, SRR5047988, SRR5047989, SRR5047990, SRR5047991, SRR5047992, SRR5047993, SRR5047994, SRR5171100 Páncreas SRR5171086, SRR5171087 Colon sigmoide SRR5171098, SRR5171099 Intestino delgado SRR5048001, SRR5048002, SRR5048003, SRR5048004, SRR5048005, SRR5048006, SRR5048007, SRR5048008, SRR5048009, SRR5048010, SRR171080, SRR5171081 Bazo SRR5047941, SRR5047942, SRR5047943, SRR5047944, SRR5047945, SRR5047946, SRR5171241, SRR5171242 Estómago SRR5047997, SRR5047996, SRR5047995, SRR5047998, SRR5048000, SRR5047999 Tejido adiposo subcutáneo SRR5048011, SRR5048012, SRR5048013, SRR5048014 Testículos SRR5047953, SRR5047954, SRR5047955, SRR5047956, SRR5171085, SRR5171084 Timo SRR5047947, SRR5047948, SRR5047949, SRR5047950, SRR5047951, SRR5047952 Vejiga urinaria SRR5048035, SRR5048036 Table 6: Accession numbers of the GTEx data sets used in this study Tejido Números de acceso (SRA) de donantes seleccionados al azar Adiposo Subcutáneo SRR599313 SRR608150 SRR608198 SRR612263 SRR612707 SRR612815 SRR612863 SRR612935 SRR613150 SRR613234 SRR613342 SRR613390 SRR613533 SRR613550 SRR613639 SRR613675 SRR613855 SRR613896 SRR613915 SRR613927 SRR614119 SRR614191 SRR614395 SRR614419 SRR614864 SRR615069 SRR615237 SRR615431 SRR615659 SRR615778 cccznn / i znz / B / v SRR615874 SRR615946 SRR617841 SRR654730 SRR654862 SRR655182 SRR655531 SRR655637 SRR655768 SRR655816 SRR656946 SRR657599 SRR657949 SRR658081 SRR658754 SRR658953 SRR659109 SRR654898 SRR656059 SRR658941 Glándula suprarrenal SRR1069421 SRR1070913 SRR1072626 SRR1073365 SRR1073775 SRR1074474 SRR1075314 SRR1076632 SRR1076823 SRR1O82O35 SRR1082616 SRR1082733 SRR1083824 SRR1083892 SRR1085590 SRR1085951 SRR1086046 SRR1087297 SRR1087511 SRR1087606 SRR1088365 SRR1088461 SRR1089479 SRR1089950 SRR1091476 SRR1092160 SRR1092329 SRR1092686 SRR1093625 SRR1093721 SRR1093954 SRR1094144 SRR1099378 SRR1099427 SRR1099598 SRR1099694 SRR1100496 SRR1100728 SRR808862 SRR809873 SRR810129 SRR810713 SRR811237 SRR811631 SRR812246 SRR814407 SRR816495 SRR816865 SRR817649 SRR818694 Arteria Aorta SRR1069376 SRR1072749 SRR1075579 SRR1079998 SRR1082283 SRR1084460 SRR808836 SRR810201 SRR813632 SRR819293 SRR1070111 SRR1070641 SRR1071644 SRR1072078 SRR1073705 SRR1074478 SRR1074622 SRR1075028 SRR1076343 SRR1077090 SRR1078586 SRR1079023 SRR1080148 SRR1081137 SRR1081519 SRR1081910 SRR1083076 SRR1083286 SRR1083604 SRR1084276 SRR1085159 SRR654850 SRR808044 SRR808152 SRR8O8351 SRR808914 SRR809320 SRR809470 SRR809785 SRR809831 SRR81O367 SRR811333 SRR8U471 SRR811819 SRR812673 SRR815092 SRR816565 SRR817744 SRR818232 SRR818999 Vejiga SRR1071717 SRR1092208 SRR2135324 SRR1079830 SRR1093930 SRR2135407 SRR1081765 SRR1097296 SRR1085402 SRR1099957 SRR1086236 SRR1120296 Cerebro Corteza SRR1081741 SRR1310136 SRR1314958 SRR1323043 SRR1082262 SRR1311400 SRR1315269 SRR1323746 SRR1O83632 SRR1311575 SRR1315866 SRR1324371 SRR1085975 SRR1311794 SRR1316815 SRR1327593 SRR1310008 SRR1312428 SRR1320280 SRR1328487 cccznn / i ζπζ / β / υ SRR598332 SRR601006 SRR601669 SRR602927 SRR603333 SRR604026 SRR608662 SRR612575 SRR614310 SRR615213 SRR615838 SRR627421 SRR627425 SRR627449 SRR627455 SRR654874 SRR656745 SRR659555 SRR660626 SRR660933 SRR663320 SRR663753 SRR664854 SRR808614 SRR810319 SRR810877 SRR812012 SRR812436 SRR816770 SRR820078 SRR1068977 SRR1068999 SRR 1070208 SRR1070260 SRR1070738 SRR1071084 SRR1071905 SRR1074860 SRR1075484 SRR1076219 SRR1076441 SRR1077139 SRR1077920 SRR1078258 SRR1079948 SRR1081023 SRR1082859 SRR1083052 SRR1083959 SRR1084079 SRR1084674 SRR1086538 SRR1086772 SRR615910 SRR655447 SRR655852 SRR656911 SRR656970 SRR657018 SRR657528 SRR658105 SRR658319 Mama SRR658409 SRR659223 SRR660248 SRR660283 SRR662306 SRR662378 Tejido SRR662811 SRR808428 SRR808942 SRR811073 SRR811285 SRR812198 mamario SRR813868 SRR815208 SRR816336 SRR818873 SRR820571 SRR821498 Cérvix SRR1075223 SRR1088832 SRR1089562 SRR1096876 SRR1097035 Ectocérvix SRR1097574 SRR1069943 SRR1074337 SRR1077380 SRR1081068 SRR1O835O4 SRR1083678 SRR1084505 SRR1086020 SRR1087271 SRR1090431 SRR1091524 SRR1092493 SRR1093366 SRR1102198 SRR1102224 SRR1102998 SRR1308269 SRR1312577 SRR1312666 SRR1312784 SRR1317110 SRR1317653 SRR1318624 SRR1319038 SRR1320445 SRR1320490 SRR1321377 SRR 1322070 SRR 1323002 SRR1323215 SRR1324473 SRR1327454 SRR1327505 SRR1327527 SRR1327570 SRR1328528 SRR1328980 SRR1329642 SRR1329663 SRR1330176 Colon SRR1330770 SRR133O831 SRR1332467 SRR1333167 SRR1333287 Sigmoide SRR1334011 SRR1334055 SRR1334181 SRR1336617 SRR1336863 SRR1069231 SRR1069255 SRR1069328 SRR 1069666 SRR1069871 SRR1070036 SRR1070060 SRR1070620 SRR1070665 SRR1071207 SRR1071499 SRR1072055 SRR1072297 SRR1072388 SRR1072480 Esófago SRR1073631 SRR1074450 SRR1074502 SRR1074578 SRR1075458 Mucosa SRR1075603 SRR1076195 SRR1076705 SRR1076801 SRR1077310 SRR1077356 SRR1077619 SRR1077850 SRR1078140 SRR1078538 SRR807679 SRR807703 SRR809406 SRR809919 SRR812294 SRR812318 SRR813283 SRR813505 SRR813536 SRR814467 SRR815116 SRR815568 SRR816403 SRR817306 SRR819124 SRR819559 SRR819637 SRR820280 SRR820689 SRR821282 Trompa de Falopio SRR1071359 SRR1074140 SRR1076584 SRR1082520 SRR1083776 SRR1101693 SRR811938 Corazón- Ventrículo izquierdo SRR598148 SRR598509 SRR598589 SRR599025 SRR599086 SRR599249 SRR599380 SRR600474 SRR600829 SRR600852 SRR600924 SRR601239 SRR601613 SRR601645 SRR601868 SRR601986 SRR602106 SRR602437 SRR602461 SRR603449 SRR603918 SRR603968 SRR604122 SRR604174 SRR604206 SRR604230 SRR606939 SRR607252 SRR607313 SRR607970 SRR608096 SRR608480 SRR612335 SRR612719 SRR612875 SRR613186 SRR613462 SRR613510 SRR613759 SRR614215 SRR614683 SRR614996 SRR615335 SRR615359 SRR615898 SRR615970 SRR655792 SRR657903 SRR658283 SRR658331 Riñón Corteza SRR1071807 SRR1080366 SRR1085759 SRR1089504 SRR1105272 SRR1314940 SRR1317086 SRR1325483 SRR1328447 SRR1329154 SRR1340662 SRR1362263 SRR1377578 SRR138O931 SRR1396700 SRR1416516 SRR1420649 SRR1432650 SRR1433066 SRR1435730 SRR1437274 SRR1442708 SRR1443092 SRR1445835 SRR1447631 SRR1452888 SRR1456711 SRR1465871 SRR1468426 SRR1469746 SRR1486080 SRR1490658 SRR1500261 SRR2135353 SRR2135396 SRR809943 SRR810007 SRR821356 Hígado SRR1069141 SRR1070689 SRR1071668 SRR1073435 SRR1075102 SRR1075804 SRR1076022 SRR1080117 SRR1080294 SRR1081184 SRR1082151 SRR1083983 SRR1086256 SRR1087007 SRR1087321 SRR1089446 SRR1090095 SRR1090556 SRR1091865 SRR1093861 SRR1095383 SRR1095913 SRR1098737 SRR1100991 SRR1101883 SRR1102152 SRR1102899 SRR1105248 SRR1120939 SRR1310433 SRR1312266 SRR1313807 SRR1316096 SRR13175372 SRR15 SRR1321877 SRR1322312 SRR1322477 SRR1323491 SRR1324295 SRR1324412 SRR1325290 SRR1328760 SRR1331488 SRR1334866 SRR1335236 SRR1336314 SRR815140 SRR815711 SRR821043 Pulmón SRR1070015 SRR1070358 SRR1071568 SRR1072150 SRR1073119 SRR1074769 SRR1081283 SRR1084602 SRR1084766 SRR1086728 SRR1087559 SRR1091670 SRR1095695 SRR1098074 SRR1098785 SRR1098998 SRR1099286 SRR1099546 SRR1102079 SRR1102804 SRR1307123 SRR1307615 SRR1308239 SRR1308504 SRR1308939 SRR1309452 SRR1309468 SRR1309490 SRR131O313 SRR1310520 SRR1310797 SRR1310959 SRR1310975 SRR1312209 SRR1312522 SRR1312558 SRR813043 SRR814244 SRR814703 SRR817004 SRR817070 SRR817166 SRR817488 SRR818499 SRR819186 SRR819318 SRR819658 SRR820596 SRR821302 SRR821525 Glándula salival menor SRR1071105 SRR1078392 SRR1080790 SRR1081589 SRR1097245 SRR1100608 SRR1315412 SRR1318089 SRR1321897 SRR1325201 SRR1328715 SRR1330723 SRR1331771 SRR1338384 SRR1339987 SRR1340260 SRR1348929 SRR13536OO SRR1356057 SRR1358391 SRR1376380 SRR1376450 SRR1376741 SRR1381185 SRR1382978 SRR1385690 SRR1386927 SRR1388459 SRR1389955 SRR1397720 SRR1400931 SRR1404339 SRR1405147 SRR1406135 SRR1406348 SRR1407044 SRR1413307 SRR1416141 SRR1416188 SRR1416841 SRR1418225 SRR1418473 SRR1418747 SRR1419561 SRR1429429 SRR1429540 SRR1431823 SRR1432868 SRR1432958 SRR1433493 Músculo- Esqueleto SRR1068855 SRR1071231 SRR1071594 SRR1071955 SRR1074359 SRR1074670 SRR1074719 SRR1077288 SRR1077805 SRR1080766 SRR1084369 SRR1084417 SRR1085519 SRR1087245 SRR1087825 SRR1088581 SRR1089424 SRR1089901 SRR1090265 SRR1092349 SRR1092985 SRR1094051 SRR1095720 SRR1096174 SRR1096662 SRR1098474 SRR1098879 SRR1100588 SRR1102830 SRR1105057 SRR812773 SRR813656 SRR8138O2 SRR813983 SRR815020 SRR815044 SRR815470 SRR815783 SRR815825 SRR816015 SRR816226 SRR816382 SRR817282 SRR817421 SRR8186OO SRR818773 SRR818901 SRR819054 SRR819261 SRR820907 Nervio Tibial SRR1070086 SRR1070159 SRR1070597 SRR1072724 SRR1073553 SRR1074550 SRR1075384 SRR1075825 SRR1076559 SRR1079636 SRR1079850 SRR1080093 SRR1082059 SRR1082809 SRR1086417 SRR1087079 SRR1088706 SRR1090070 SRR1091184 SRR1092062 SRR1095334 SRR1096007 SRR1096222 SRR1096478 SRR1096500 SRR1096806 SRR1097055 SRR1098385 SRR1310455 SRR1310645 SRR1311131 SRR1311308 SRR1312370 SRR1312464 SRR813704 SRR814052 SRR814996 SRR815422 SRR815685 SRR817026 SRR817397 SRR817539 SRR817609 SRR818939 SRR818961 SRR820350 SRR820402 SRR821096 SRR821124 SRR821255 Ovario SRR1071475 SRR1073389 SRR1073878 SRR1075360 SRR1078042 SRR1078636 SRR1078735 SRR1081987 SRR1082352 SRR1082471 SRR1085565 SRR1085736 SRR1086212 SRR1086656 SRR1088856 SRR1089134 SRR1090698 SRR1090928 SRR1091164 SRR1092038 SRR1093601 SRR1093747 SRR1096458 SRR1097124 SRR1097148 SRR1098807 SRR1099310 SRR1099669 SRR1101453 SRR1101859 SRR1102005 SRR1102780 SRR1120276 SRR1312446 SRR1315495 SRR1316513 SRR1319793 SRR1336244 SRR1339699 SRR1340598 SRR1341583 SRR1342849 SRR1347518 SRR1350891 SRR1351641 SRR1353537 SRR814293 SRR814892 SRR816629 SRR821072 Páncreas SRR1069352 SRR1070403 SRR1070764 SRR1071519 SRR1072007 SRR1072104 SRR1072972 SRR1073021 SRR1073167 SRR1073991 SRR1074090 SRR1074385 SRR1075174 SRR1075336 SRR1076244 SRR1076868 SRR1078066 SRR1079754 SRR1080624 SRR1082080 SRR1082544 SRR1084128 SRR1084323 SRR1085187 SRR1O8531O SRR1086070 SRR1087728 SRR1088291 SRR1088413 SRR1088537 SRR1089537 SRR1089688 SRR1091032 SRR1091144 SRR1092937 SRR1093340 SRR1093434 SRR1093577 SRR1095407 SRR1095479 SRR1095651 SRR1097777 SRR1097883 SRR812745 SRR813208 SRR816541 SRR819771 SRR821O5O SRR821231 SRR821666 Pituitaria SRR1076393 SRR1077455 SRR1077708 SRR1077968 SRR1082664 SRR1082685 SRR1089785 SRR1096101 SRR1096339 SRR1101612 SRR1309119 SRR1309638 SRR1310817 SRR1311599 SRR1311709 SRR1311958 SRR1317963 SRR1318026 SRR1319946 SRR1321650 SRR1323977 SRR1324141 SRR1324184 SRR1325161 SRR1325944 SRR1326408 SRR1326797 SRR1328143 SRR1331962 SRR1332024 SRR1332904 SRR1336029 SRR1336529 SRR1337321 SRR1339007 SRR1340241 SRR1343012 SRR1343221 SRR1343720 SRR1343778 SRR1345329 SRR1347236 SRR1347278 SRR1347389 SRR813959 SRR815920 SRR816517 SRR816609 SRR816677 SRR821573 Próstata SRR1069209 SRR1069514 SRR1073069 SRR1074410 SRR1075126 SRR1075530 SRR1076120 SRR1077429 SRR1078164 SRR1078684 SRR1078855 SRR1080318 SRR1080696 SRR1081789 SRR1082496 SRR1083732 SRR1086441 SRR1086514 SRR1086869 SRR1091645 SRR1091990 SRR1092444 SRR1092468 SRR1092636 SRR1092913 SRR1093075 SRR1093697 SRR1096081 SRR1097344 SRR1098686 SRR1099402 SRR1105441 SRR13O886O SRR1310939 SRR1312002 SRR1315353 SRR1317751 SRR1323699 SRR1324314 SRR1326100 SRR1332360 SRR1335605 SRR1335964 SRR8131O8 SRR815280 SRR815542 SRR815845 SRR816818 SRR816969 SRR820234 Piel - No expuesta al sol (suprapúbica) SRR1069048 SRR1070232 SRR1070888 SRR1073605 SRR1074289 SRR1075247 SRR1076292 SRR1077263 SRR1077898 SRR1079434 SRR1083215 SRR1083579 SRR1084299 SRR1087801 SRR1091597 SRR1094216 SRR1095503 SRR1096408 SRR1098216 SRR1100703 SRR1309920 SRR1309985 SRR1310053 SRR1311153 SRR1311224 SRR1311916 SRR1312124 SRR1312244 SRR1312645 SRR1312934 SRR1313494 SRR1314036 SRR1314137 SRR1314728 SRR1314810 SRR1315912 SRR1316438 SRR1316747 SRR1316833 SRR1317022 SRR814491 SRR815164 SRR815350 SRR815759 SRR815805 SRR818372 SRR818440 SRR819844 SRR820427 SRR820810 Intestino delgado íleon terminal SRR1070133 SRR1071181 SRR1072602 SRR1074934 SRR1076046 SRR1076465 ​​SRR1077728 SRR1079973 SRR1084154 SRR1085378 SRR1087680 SRR1310497 SRR1311731 SRR1313664 SRR1319059 SRR1319301 SRR1321483 SRR1326449 SRR1326845 SRR1329508 SRR1330371 SRR1337749 SRR1337930 SRR1338402 SRR1339086 SRR1340762 SRR1340782 SRR1343136 SRR1344079 SRR1344364 SRR1351907 SRR1354400 SRR1356327 SRR13588O3 SRR1359027 SRR1359587 SRR1360321 SRR1361391 SRR1365655 SRR1365767 SRR1366102 SRR1366412 SRR1367520 SRR1375371 SRR1378199 SRR1379036 SRR1380358 SRR1380436 SRR1384312 SRR1387745 Estómago SRR1068953 SRR1069166 SRR1069714 SRR1069778 SRR1070382 SRR1070549 SRR1070884 SRR1071761 SRR1072199 SRR1072700 SRR1072821 SRR1072920 SRR1073459 SRR1074066 SRR1075874 SRR1076268 SRR1076417 SRR1076990 SRR1078090 SRR1078759 SRR1079900 SRR1080672 SRR1081092 SRR1081235 SRR1081717 SRR1081935 SRR1082933 SRR1082957 SRR1083149 SRR1083191 SRR1083262 SRR1O8336O SRR1083408 SRR1084252 SRR1085450 SRR1087101 SRR1088068 SRR1088117 SRR808542 SRR810689 SRR810829 SRR811193 SRR812152 SRR813234 SRR814195 SRR814268 SRR814820 SRR815326 SRR815970 SRR819719 Testículos SRR1068788 SRR1068905 SRR1069734 SRR1070479 SRR1071379 SRR1071429 SRR1072845 SRR1073531 SRR1075607 SRR1076490 SRR1077753 SRR1078299 SRR1078612 SRR1079455 SRR1079612 SRR1080022 SRR1080811 SRR1080859 SRR1081357 SRR1081401 SRR1081449 SRR1081614 SRR1081663 SRR1081688 SRR1082307 SRR1083554 SRR1084347 SRR1087055 SRR1087535 SRR1088241 SRR1308288 SRR1309425 SRR1311329 SRR1312288 SRR1314014 SRR807517 SRR808065 SRR809667 SRR81O531 SRR810899 SRR811447 SRR812912 SRR813431 SRR814082 SRR814943 SRR815588 SRR817512 SRR818850 SRR820839 SRR821518 Thyroid SRR597952 SRR598068 SRR598100 SRR598364 SRR598565 SRR598645 SRR599122 SRR599346 SRR599412 SRR601157 SRR601359 SRR601525 SRR601549 SRR601843 SRR601962 SRR602338 SRR602389 SRR602951 SRR602978 SRR603036 SRR603268 SRR603726 SRR603834 SRR603942 SRR604148 SRR604294 SRR604342 SRR607502 SRR607679 SRR607705 SRR608064 SRR608120 SRR608512 SRR613018 SRR613258 SRR613402 SRR613711 SRR613795 SRR613975 SRR614023 SRR614107 SRR614275 SRR614743 SRR614912 SRR615285 SRR615347 SRR615491 SRR615886 SRR654969 SRR655696 Útero SRR1069466 SRR1071737 SRR1073483 SRR1074430 SRR1075850 SRR1077159 SRR1077211 SRR1077996 SRR1078114 SRR1078188 SRR1078212 SRR1079213 SRR1079408 SRR1079874 SRR1080342 SRR1082128 SRR1084553 SRR1O85358 SRR1086369 SRR1309745 SRR1313991 SRR1319242 SRR1319991 SRR1321720 SRR1323234 SRR1329423 SRR1330082 SRR1336682 SRR1338468 SRR1339258 SRR1343943 SRR1353686 SRR1358126 SRR1360280 SRR1361138 SRR1361838 SRR1363718 SRR1374543 SRR1381372 SRR1382780 SRR1383237 SRR1387132 SRR1388257 SRR808704 SRR81O1O5 SRR815256 SRR817817 SRR818139 SRR818646 SRR820026 Table 7a: Information on the samples used in this study - murine samples Name of the sample Type of biological sample of Strain H-2- H-2-D K H-2- L Type of replication Cells Nb C57BL / EL4 cell line cells 6 C57BL / b b not replicated 5 000 000 mTEChi_l primary cells 6 C57BL / b b - biological 51 237 mTEChi_2 primary cells 6 C57BL / b b - biological 31 686 mTEChi_3 primary 6 b b - biological 31 702 CT26 cell line Balb / c cells d d d not replicated 5 000 000 mTEChi_l primary Balb / c cells d d d biological 16 338 mTEChi_2 primary Balb / c cells d d d biological 19 782 mTEChi_3 primary Balb / c d d d biological 23 130 cccznn / i ζηζ / ε / γ Sample BioAnalyser_ Name RIN Input RNA_n g Nucleic Type Strand Acid Specificity Platform ReadType EL4 9.95 4000 HiSeq 2000 Strand-Specific Polyadenylated mRNA Paired End mTEChi_l 10 100 HiSeq 2000 Strand-Specific Polyadenylated mRNA Paired End mTEChi_2 9 .2 100 HiSeq 2000 Strand-Specific Polyadenylated mRNA Paired End mTEChi_3 9.9 100 HiSeq 2000 Strand-Specific Polyadenylated mRNA Paired End CT26 10 2000 HiSeq 2000 Strand-Specific Polyadenylated mRNA Paired End mTEChi_l 9.5 50 NexSeq Strand-Specific Polyadenylated mRNA 500 Paired End mTEChi_2 9.4 50 Strand-Specific Polyadenylated mRNA NextSeq 500 Paired End mTEChi_3 9.2 50 Strand-Specific Polyadenylated mRNA NextSeq 500 Paired End Reading Name Readings Number of NbRepM NbCellsM Code of the sample Length_bp Total Nb Access_GEO S S (xlO Λ 6) access_ data MS EL4 100 240372644 GSE111092 3 250 PXD009064 mTEChi_l 100 159208840 GSE111092 N / A N / A N / A mTEChi_2 100 145643202 GSE111092 N / A N / A N / A mTEChi_3 100 152139924 GSE111092 N / A N / A N / A CT26 100 247522370 GSE111092 3 250 PXD009065 mTEChi_l 80 156128844 GSE111092 N / A N / A N / A mTEChi_2 80 161566962 GSE111092 N / A N / A N / A mTEChi 3 80 137929352 GSE111092 N / A N / A N / A Table 7b: Information on the samples used in this study - human samples Name of the sample Type of biological sample HLA-A HLA-B HLA- C Type of replication Cells Nb 07H103 primary leukemic cells 01:01 | 02:01 40:01 | 44:02 03:04 | 05:01 r a not replicated 2,600,000 10H080 primary leukemia cells 02:01 | 11:01 40:01 | 44:03 03:04 | 16:01 not replicated 2,000,000 10H118 primary leukemic cells 01:01 | 02:01 07:02 | 08:01 07:01 | 07:17 1' a unreplicated 3,400,000 12H018 primary leukemic cells 02:01 | 11:01 07:02 | 35:03 07:02 | 12:03 r a not replicated 4,000,000 lc2 tumor biopsy 11:01 | 23:01 35:01 | 44:03 04:01 not replicated N / A lc4 tumor biopsy 02:01 | 03:01 07:02 07:02 not replicated N / A lc6 01:01 biopsy | 08:01 | 02:02 | not replicated N / A tumor 24:02 27:13 07:01 102015 primary TEC N / A N / A N / A not replicated 33 076 062015 primary TEC N / A N / A N / A not replicated 84 198 S5 primary mTEC N / A N / A N / A not replicated 59 197 S9 primary mTEC N / A N / A N / A not replicated 100 719 SO mTEC primary N / A N / A N / A not replicated 50 058 Sil mTEC primary N / A N / A N / A not replicated 100 506 Come in Sample Name BioAnalyser _RIN a RNA_n g Nucleic Acid Type Strand Specific Platform ReadTyp e 07H103 10,500 Strand-Specific Polyadenylated mRNA HiSeq 2000 Paired End 10H080 10,500 HiSeq 2000 Strand-Specific Polyadenylated mRNA Paired End 10H1018 9 5 RNA HiSeq 2000 strand-specific polyadenylated Paired end 12H018 9500 HiSeq 2000 strand-specific polyadenylated mRNA Paired end lc2 9.2 4000 HiSeq 2000-specific mRNA End polyadenylated paired-strand lc4 9.4 4000 polyadenylated strand-specific mRNA HiSeq 2000 Paired-end lc6 8.9 4000 polyadenylated-strand-specific mRNA HiSeq 2000 Paired-end 102015 7 8 polyadenylated paired-end mRNA NextSeq 500 Paired-end 06201 7 13 Strand-specific Polyadenylated mRNA NextSeq 500 Paired End S5 7 41 Strand-specific Polyadenylated mRNA NextSeq 500 Paired End S9 8 56 Strand-specific Polyadenylated mRNA NextSeq 500 Paired End SO 8 68 Strand-specific Polyadenylated mRNA NextSeq 500 Paired End c Sil 7 59 strand-specific polyadenylated mRNA NextSeq 500 Paired end Name Code of Reading Readings Nb Number of NbRepM NbCellsM of the access_ Total Length_bp GEO_Access S S (xlOΛ6) shows MS data 07H103 100 105 863 640 GSE113972 3 650 PXD009749 10H080 100 129 444 492 GSE113972 3 / 4 500 / 100 PXD009753 / PXD007935 10H118 100 226 508 070 GSE113972 3 700 PXD009750 12H018 100 161 724 658 GSE113972 3 465 PXD009751 lc2 100 268 396 930 GSE113972 2 N / A PXD009752 lc4 100 262 531 548 GSE113972 2 N / A PXD009754 LC6 100 246 868 078 GSE113972 2 N / A PXD009755 102015 80 134 624 214 N / A Specific HEBRA NEXTSEQ 500 Extreme Parent 532 N / A Strand Specific NextSeq 500 Paired End S9 80 229 281 098 N / A Strand Specific NextSeq 500 Strand Specific SO 80 231 185 678 N / A Strand Specific NextSeq 500 Paired End Sil 80 251 770 122 N / A Specific NextSeq 500 Strand Paired End cccznn / i znz / B / v Although the present invention has been described above by means of specific embodiments thereof, it may be modified, without departing from the spirit and nature of the invention in question as defined in the appended claims. In the claims, the word comprising is used as an open-ended term, substantially equivalent to the phrase including, but not limited to. The singular forms "a", "an" and "the" include the corresponding references in the plural unless the context clearly dictates otherwise. REFERENCES 1. Mlecnik, B. et al., The tumor microenvironment and immunoscore are critical determinants of dissemination to distant metastasis. Sel Transí Med 8, 327ra326 (2016). 2. ChaiOentong, P. et al., Pan-cancer immunogenomic analyzes reveal genotypeimmunophenotype relationships and predictors of response to checkpoint blockade. Cell Rep 18,248-262 (2017). 3. Shao, W. et al., The systeMHC Atlas project. Nucleic Acids Res 46, D1237D1247 (2018). 4. Martin, S.D., Coukos, G., Holt, R.A. & Nelson, B.H. Targeting the undruggable: Immunotherapy meets personalized oncology in the genomic era. Ann Oncol 26, 2367-2374 (2015). 5. Marty, R. et al., MHC-I genotype restricts the oncogenic mutational landscape. Cell 171, 1272-1283 el215 (2017). 6. Zhong, S. et al., T-cell receptor affinity and avidity defines antitumor response and autoimmunity in T-cell immunotherapy. Proc Nati Acad Sci USA 110, 6973-6978 (2013). 7. Sahin, U. et al., Personalized RNA mutanome vaccines mobilize poly-specific therapeutic immunity against cancer. Nature 547, 222-226 (2017). 8. Turajlic, S. et al., Insertion-and-deletion-derived tumor-specific neoantigens and the immunogenic phenotype: a pan-cancer analysis. Lancet Oncol 18, 1009-1021 (2017). 9. Yadav, M. et al., Predicting immunogenic tumor mutations by combining mass spectrometry and exorne sequencing. Nature 515, 572-576 (2014). 10. Pearson, H. et al., MHC class I-associated peptides derive from selective regions of the human genome. J Clin Invest 126, 4690-4701 (2016). 11. Tran, E. et al., Immunogenicity of somatic mutations in human gastrointestinal cancers. Science 350, 1387-1390 (2015). 12. Gros, A. et al., Prospective Identification of neoantigen-specific lymphocytes in the peripheral blood of melanoma patients. NatMed 22, 433-438 (2016). 13. Bassani-Sternberg, M. et al., Direct Identification of clinically relevant neoepitopes presented on native human melanoma tissue by mass spectrometry. Nat Commun 7, 13404 (2016). 14. Mertens, F., Johansson, B., Fioretos, T. & Mitelman, F. The emerging complexity of gene fusions in cancer. Nat Rev Cancer 15, 371-381 (2015). 15. Baca, S.C. et al., Punctuated evolution of prostate cancer genomes. Cell 153, 666677 (2013). 16. Hayward, N.K. et al., Whole-genome landscapes of major melanoma subtypes. Nature 545, 175-180 (2017). 17. Khurana, E. et al., Role of non-coding sequence variants in cancer. Nat Rev Genet 17, 93-108 (2016). 18. Laumont, C.M. et al., Global proteogenomic analysis of human MHC class I-associated peptides derived from non-canonical reading frames. Nat Commun 7, 10238 (2016). 19. Rooney, M.S., Shukla, S.A., Wu, C.J., Getz, G. & Hacohen, N. Molecular and genetic properties of tumors associated with local immune cytolytic activity. Cell 160, 4861 (2015). 20. Charoentong, P. et al., Pan-cancer Immunogenomic Analyzes Reveal Genotype Immunophenotype Relationships and Predictors of Response to Checkpoint Blockade. Cell Rep 18, 248-262 (2017). 21. Anwar, S.L., Wulaningsih, W. & Lehmann, U. Transposable Elements in Human Cancer: Causes and Consequences of Deregulation. Int JMol Sci 18(2017). 22. Kassiotis, G. & Stoye, J.P. Immune responses to endogenous retroelements: taking the bad with the good. Nat Rev Immunol 16, 207-219 (2016). 23. Kershaw, M.H. et al., Immunization against endogenous retroviral tumor-associated antigens. Cancer Res 61, 7920-7924 ​​(2001). 24. Sacha, J.B. et al., Vaccination with cancer- and HIV infection-associated endogenous retrotransposable elements is safe and immunogenic. J Immunol 189, 14671479 (2012). 25. Malarkannan, S., Serwold, T., Nguyen, V., Sherman, L.A. & Shastri, N. The mouse mammary tumor virus env gene is the source of a CD8+ T-cell-stimulating peptide presented by a major histocompatibility complex class I molecule in a murine thymoma. Proe Nati Acad Sci USA 93, 13991-13996 (1996). 26. Huang, A.Y. et al., The immunodominant major histocompatibility complex class I-restricted antigen of a murine colon tumor derives from an endogenous retroviral gene product. Proc Nati Acad Sci USA 93, 9730-9735 (1996). 27. Schiavetti, F., Thonnard, J., Colau, D., Boon, T. & Coulie, P.G. A human endogenous retroviral sequence encoding an antigen recognized on melanoma by cytolytic T lymphocytes. Cancer Res 62, 5510-5516 (2002). 28. Takahashi, Y. et al., Regression of human kidney cancer following allogeneic stem cell transplantation is associated with recognition of an HERV-E antigen by T cells. J Clin Invest 118, 1099-1109 (2008). 29. Kim, M.J., Miller, C.M., Shadrach, J.L., Wagers, A.J. & Serwold, T. Young, proliferative thymic epithelial cells engraft and function in aging thymuses. J Immunol 194, 4784-4795 (2015). 30. Dobin, A. et al., STAR: ultrafast universal RNA-seq aligner. Bioinformatics 29, 15-21 (2013). 31. Quinlan, A.R. & Hall, I.M. BEDTools: a flexible suite of Utilities for comparing genomic features. Bioinformatics 26, 841-842 (2010). 32. Carón, E. et al., The MHC I immunopeptidome conveys to the cell surface an integrative view of cellular regulation. Mol Syst Biol 7, 533 (2011). 33. Andreatta, M. & Nielsen, M. Gapped sequence alignment using artificial neural networks: application to the MHC class I system. Bioinformatics 32, 511-517 (2016). 34. Robinson, J.T. et al, Integrative genomics viewer. Nat Biotechnol 29, 24-26 (2011). 35. Bereman, M.S. et al., An Automated Pipeline to Monitor System Performance in Liquid Chromatography-Tandem Mass Spectrometry Proteomic Experiments. J Proteome Res 15, 4763-4769 (2016). 36. Yue, F. et al., A comparative encyclopedia of DNA elements in the mouse genome. Nature 515, 355-364 (2014). 37. Barbosa-Morais, N.L. et al., The evolutionary landscape of alternative splicing in vertebrate species. Science 338, 1587-1593 (2012). 38. Patenaude, J. & Perreault, C. Thymic Mesenchymal Cells Have a Distinct Transcriptomic Profile. J Immunol 196, 4760-4770 (2016). 39. St-Pierre, C., Trofimov, A., Brochu, S., Lemieux, S. & Perreault, C. Differential Features of AIRE-Induced and AIRE-Independent Promiscuous Gene Expression in Thymic Epithelial Cells. J Immunol 195, 498-506 (2015). 40. Dumont-Lagace, M., St-Pierre, C. & Perreault, C. Sex hormones have pervasive effects on thymic epithelial cells. Sci Rep 5, 12895 (2015). 41. Dumont-Lagace, M., Brochu, S., St-Pierre, C. & Perreault, C. Adult thymic epithelium contains nonsenescent label-retaining cells. J Immunol 192, 2219-2226 (2014). 42. deVerteuil, D.A. et al., Immunoproteasomes shape the transcriptome and regulate the function of dendritic cells. J Immunol 193, 1121-1132 (2014). 43. de Verteuil, D. et al., Deletion of immunoproteasome subunits imprints on the transcriptome and has a broad impact on peptides presented by major histocompatibility complex I molecules. Mol Cell Proteomics 9, 2034-2047 (2010). 44. Moon, LJ. et al., Naive CD4(+) T cell frequency varies for different epitopes and predicts repertoire diversity and response magnitude. Immunity 27, 203-213 (2007). 45. Legoux, F.P. & Moon, J.J. Peptide: MHC tetramer-based enrichment of epitopespecific T cells. J Vis Exp (2012). 46. ​​McFarland, H.I., Nahill, S.R., Maciaszek, J.W. & Welsh, R.M. CDllb (Mac-1): a marker for CD8+ cytotoxic T cell activation and memory in virus infection. J Immunol 149, 1326-1333 (1992). 47. Chadbum, A., Inghirami, G. & Knowles, D.M. Hairy cell leukemia-associated antigen LeuM5 (CDllc) is preferentially expressed by benign activated and neoplastic CD8 T cells. Am J Pathol 136, 29-37 (1990). 48. Nesvizhskii, A.I. Proteogenomics: concepts, applications and computational strategies. Nat Methods 11, 1114-1125 (2014). 49. Noble, W.S. Mass spectrometrists should search only for peptides they care about. Nat Methods 12, 605-608 (2015). 50. Murphy, J.P. et al., MHC-I Ligand Discovery Using Targeted Database Searches of Mass Spectrometry Data: hnplications for T-Cell Immunotherapies. J Proteome Res 16, 1806-1816(2017). 51. Granados, D.P. et al., Impact of genomic polymorphisms on the repertoire of human MHC class I-associated peptides. Nat Commun 5, 3600 (2014). 52. Bassani-Sternberg, M., Pletscher-Frankild, S., Jensen, L.J. & Mann, M. Mass spectrometry of human leukocyte antigen class I peptidomes reveáis strong effects of protein abundance and turnover on antigen presentation. Mol Cell Proteomics 14, 658-673 (2015). 53. Fortier, M.H. et al., The MHC class I peptide repertoire is molded by the transcriptome. J Exp Med 205, 595-610 (2008). 54. Jenkins, M.K. & Moon, J.J. The role of naive T cell precursor frequency and recruitment in dictating immune response magnitude. J Immunol 188, 4135-4140 (2012). 55. Obar, J.J., Khanna, K.M. & Lefrancois, L. Endogenous naive CD8+ T cell precursor frequency regulates primary and memory responses to infection. Immunity 28, 859-869 (2008). 56. The Grotto, N.L. et al., Primary CTL response magnitude in mice is determined by the extent of naive T cell recruitment and subsequent clonal expansion. J Clin Invest 120, 1885-1894 (2010). 57. Mueller, S.N., Gebhardt, T., Carbone, F.R. & Heath, W.R. Memory T cell subsets, migration patterns, and tissue residence. Annu Rev Immunol 31, 137-161 (2013). 58. Baaten, B.J., Tinoco, R., Chen, A.T. & Bradley, L.M. Regulation of Antigen Experienced T Cells: Lessons from the Quintessential Memory Marker CD44. Front Immunol3, 23 (2012). 59. Laugel, B. et al., Different T cell receptor affinity thresholds and CD8 coreceptor dependence govern cytotoxic T lymphocyte activation and tetramer binding properties. J Biol Chem 282, 23799-23810 (2007). 60. Richards, D.M., Kyewski, B. & Feuerer, M. Re-examining the Nature and Function of Self-Reactive T cells. Trends Immunol 37, 114-125 (2016). 61. McGranahan, N. et al., Clonal neoantigens elicit T cell immunoreactivity and sensitivity to immune checkpoint blockade. Science 351, 1463-1469 (2016). 62. Assarsson, E. et al., A quantitative analysis of the variables affecting the repertoire of T cell specificities recognized after vaccinia virus infection. J Immunol 178, 7890-7901 (2007). 63. Martin, S.D. et al., Low Mutation Burden in Ovarian Cancer May Limit the Utility of Neoantigen-Targeted Vaccines. PLoS One 11, e0155189 (2016). 64. Rudensky, A., Preston-Hurlburt, P., Hong, S.C., Barlow, A. & Janeway, C.A., Jr. Sequence analysis of peptides bound to MHC class II molecules. Nature 353, 622-627 (1991). 65. Meydan, C., Otu, H.H. & Sezerman, O.U. Prediction of peptides binding to MHC class I and II allies by temporal motif mining. BMC Bioinformatics 14 Supplement 2, S13 (2013). 66. Szpakowski, S. et al., Loss of epigenetic silencing in tumors preferentially affects primate-specific retroelements. Gene 448, 151-167 (2009). 67. Capietto, A.H., Jhunjhunwala, S. & Delamarre, L. Characterizing neoantigens for personalized cancer immunotherapy. Curr Opin Immunol 46, 58-65 (2017). 68. Helft, J. et al., GM-CSF Mouse Bone Marrow Cultures Comprise a Heterogeneous Population of CDllc(+)MHCII(+) Macrophages and Dendritic Cells. Immunity 42, 1197-1211 (2015). 69. Wimmers, F., Schreibelt, G., Skold, A.E., Figdor, C.G. & DeVries, I.J. Paradigm Shift in Dendritic Cell-Based Immunotherapy: From in vitro Generated Monocyte-Derived DCs to Naturally Circulating DC Subsets. Front Immunol 5, 165 (2014). 70. Guilliams, M. & Malissen, B. A Death Notice for In-Vitro-Generated GM-CSF Dendritic Cells? Immunity 42, 988-990 (2015). 71. Melief, C.J., van Hall, T., Arens, R., Ossendorp, F. & van der Burg, S.H. Therapeutic cancer vaccines. J Clin Invest 125, 3401-3412 (2015). 72. Guo, C. et al., Therapeutic cancer vaccines: past, present, and future. Adv Cancer Res 119, 421-475 (2013). 73. Melero, I. et al., Therapeutic vaccines for cancer: an overview of clinical trials. Nat Rev Clin Oncol 11, 509-524 (2014). 74. Baruch, E.N., Berg, A.L., Besser, M.J., Schachter, J. & Markel, G. Adoptive T cell therapy: An overview of obstacles and opportunities. Cancer 123, 2154-2162 (2017). 75. Rosenberg, S.A. & Restifo, N.P. Adoptive cell transfer as personalized immunotherapy for human cancer. Science 348, 62-68 (2015). 76. Stoeckle, C. et al., Isolation of myeloid dendritic cells and epithelial cells from human thymus. J Vis Exp, e50951 (2013). 100 77. Marcáis, G. & Kingsford, C. A fast, lock-free approach for efficient parallel counting of occurrences of k-mers. Bioinformatics 27, 764-770 (2011). the. Lanoix, J. et al., Comparison of the MHC I immunopeptidome repertoire of B-cell lymphoblasts using two isolation methods. Proteomics, el700251 (2018). 2a. Kim, M.J., Miller, C.M., Shadrach, J.L., Wagers, A.J. & Scrwold, T. Young, proliferative thymic epithelial cells engraft and function in aging thymuses. J Immunol 194, 4784-4795 (2015). 3rd Stoeckle, C. et al., Isolation of myeloid dendritic cells and epithelial cells from human thymus. J Vis Exp, e50951 (2013). 4a. Dobin, A. et al., STAR: Ultrafast universal RNA-seq aligner. Bioinformatics 29, 15-21 (2013). 5a. Daouda, T., Perreault, C. & Lemieux, S. pyGeno: A python package for precision medicine and proteogenomics. Flores 5, 381 (2016). 6a. Marcáis, G. & Kingsford, C. A fast, lock-free approach for efficient parallel counting of occurrences of k-mers. Bioinformatics 27, 764-770 (2011). 7th Carón, E. et al., The MHC I immunopeptidome conveys to the cell surface an integrative view of cellular regulation. Mol Syst Biol 7, 533 (2011). 8a. Andreatta, M. & Nielsen, M. Gapped sequence alignment using artificial neural networks: Application to the MHC class I system. Bioinformatics 32, 511-517 (2016). 9a. Robinson, J.T. et al, Integrative genomics viewer. Nat Biotechnol 29, 24-26 (2011). 10a. Yue, F. et al., A comparative encyclopedia of DNA elements in the mouse genome. Nature 515, 355-364 (2014). lia. Sloan, C.A., et al. ENCODE data at the ENCODE portal. Nucleic Acids Res 44, D726-732 (2016). 12a. Bereman, M.S. et al., An automated pipeline to monitor system performance in liquid chromatography-tandem mass spectrometry proteomic experiments. J Proteome Res 15, 4763-4769 (2016). 13th de Verteuil, D. et al., Deletion of immunopiOteasome subunits imprints on the transcriptome and has a broad impact on peptides presented by major histocompatibility complex I molecules. Mol Cell Proteomics 9, 2034-2047 (2010). 101 14 to. Vincent, K. et al., Rejection of leukemic cells requires antigen-specific T cells with high functional avidity. Biol Blood Marrow Transplant 20, 37-45 (2014). 15th Moon, J.J., et al. Naive CD4(+) T cell frequency varies for different epitopes and predicts repertoire diversity and response magnitude. Immunity 27, 203-213 (2007). 16th Legoux, F.P. & Moon, J.J. Peptide: MHC tetramer-based enrichment of epitope-specific T cells. J Vis Exp 68, 4420 (2012). 17th McFarland, H.I., Nahill, S.R., Maciaszek, J.W. & Welsh, R.M. CDllb (Mac-1): A marker for CD8+ cytotoxic T cell activation and memory in virus infection. J Immunol 149, 1326-1333 (1992). 18. Chadbum, A., Inghirami, G. & Knowles, D.M. Hairy cell leukemia-associated antigen LeuM5 (CDllc) is preferentially expressed by benign activated and neoplastic CD8 T cells. Am J Pathol 136, 29-37 (1990) 19th Vizcaino, J.A. and others, 2016 update of the PRIDE database and its related tools. Nucleic Acids Res 44, 11033 (2016). cccznn / i znz / B / v 102

Claims

1. A method for identifying a candidate tumor antigen in a sample of tumor cells, the method comprising: (a) generating a database of tumor-specific proteomes by: (i) extracting a set of subsequences (k-mers) comprising at least 33 base pairs from tumor RNA sequences; (ii) comparing the set of tumor subsequences from (i) with a set of corresponding control subsequences comprising at least 33 base pairs extracted from RNA sequences from normal cells; (iii) extracting the tumor subsequences that are absent from the corresponding control subsequences, thereby obtaining tumor-specific subsequences; and (iv) translating the tumor-specific subsequences in silico, thereby obtaining the tumor-specific proteome database;(b) generating a customized tumor proteome database by: (i) comparing tumor RNA sequences with a reference genome sequence to identify single-base mutations in said tumor RNA sequences; (ii) inserting the single-base mutations identified in (i) into the reference genome sequence, thereby creating a customized tumor genome sequence; (iii) in silico translating the protein-encoding transcripts expressed from said customized tumor genome sequence, thereby obtaining the customized tumor proteome database; (c) comparing the major histocompatibility complex (MHC)-associated peptide (MAP) sequences of said tumor with the sequences in the tumor-specific proteome database of (a) and the customized tumor proteome database of (b) to identify the MAPs;and (d) identifying a candidate tumor antigen from among the MAPs identified in (c), wherein a candidate tumor antigen is a peptide whose coding sequence ccc^nn / i ζπζ / β / υιλι 103 is overexpressed or overrepresented in Immoral cells relative to normal cells.; 2. The method of claim 1, wherein the above-mentioned method further comprises (1) isolating and sequencing the major histocompatibility complex (MHC)-associated peptides (MAPs) from the tumor cell sample, and / or (2) performing whole transcriptome sequencing on the tumor cell sample to obtain tumor RNA sequences.

3. The method of claim 2, wherein said isolation MAPs comprise (i) releasing said MAPs from said cell sample by treatment with mild acid; and (ii) subjecting the released MAPs to chromatography.

4. The method of claim 3, wherein said method further comprises filtering the released peptides with a size exclusion column prior to said chromatography.

5. The method of any of claims 1 to 4, wherein said subsequences comprise from 33 to 54 base pairs.

6. The method of any of claims 1 to 5, further comprising assembling tumor-specific subsequences overlapping into longer tumor subsequences (contigs).

7. The method of claim 6, wherein said size exclusion column has a limit of approximately 3000 Da.

8. The method of any of claims 1 to 7, wherein said MAP sequencing comprises subjecting the isolated MAPs to mass spectrometry (MS) sequencing analysis.

9. The method of any one of claims 1 to 8, wherein said method further comprises generating a personalized normal proteome database using the corresponding normal cells. cccznn / i znz / B / v 104 10. The method of claim 9, wherein said identification in (d) comprises excluding said MAP if its sequence is detected in the normal personalized proteome database.

11. The method of any one of claims 1 to 10, wherein the method further comprises generating 24- or 39-nucleotide k-mer databases from said tumor RNA sequences and normal cell RNA sequences to obtain a tumor k-mer database and a normal k-mer database; and comparing the tumor k-mer database and a normal k-mer database with 24- or 39-nucleotide k-mer derived from the sequence encoding MAP, wherein an overexpression or overrepresentation of the k-mer derived from the sequence encoding MAP in said tumor k-mer database relative to said normal k-mer database is indicative that the corresponding MAP is a candidate tumor antigen.

12. The method of claim 11, wherein the k-mer derived from the MAP coding sequence is overexpressed or overrepresented at least 10 times in said tumor k-mer database relative to said normal k-mer database.

13. The method of claim 11 or 12, wherein the k-mer derived from the MAP coding sequence is absent from said normal k-mer database.

14. The method of any one of claims 1 to 13, wherein said method comprises: (a) isolating and sequencing MAPs in a sample of tumor cells; (b) performing whole transcriptome sequencing in said sample of tumor cells, thereby obtaining tumor RNA sequences; (c) generating a database of tumor-specific proteomes by: (i) extracting a set of subsequences comprising at least 33 nucleotides from said tumor RNA sequences; (ii) comparing the set of tumor subsequences from (i) with a set of corresponding control subsequences comprising at least 33 nucleotides extracted from RNA sequences of normal cells; (iii) extracting tumor subsequences that are absent, or underexpressed by at least 4 times, in the corresponding control subsequences, thereby obtaining tumor-specific subsequences;and (iv) translating in silico the tumor-specific subsequences, thereby obtaining the tumor-specific proteome database; (d) generating a customized tumor proteome database by: (i) comparing the tumor RNA sequences with a reference genome sequence to identify single-base mutations in said tumor RNA sequences; (ii) inserting the single-base mutations identified in (i) into the reference genome sequence, thereby creating a customized tumor genome sequence; (iii) translating in silico the protein-coding transcripts expressed from said customized tumor genome sequence, thereby obtaining the customized tumor proteome database;(e) generating a customized normal proteome database by: (i) comparing RNA sequences from normal cells with a reference genome sequence to identify single-base mutations in said normal RNA sequences; (ii) inserting the single-base mutations identified in (i) into the reference genome sequence, thereby creating a customized normal genome sequence; (iii) in silico translating the protein-coding transcripts expressed from said customized normal genome sequence, thereby obtaining the customized normal proteome database; (f) generating a normal and tumor k-mer database by (i) extracting a set of subsequences comprising at least 24 nucleotides from said normal cell RNA sequences and said tumor RNA sequences;(g) comparing the MAP sequences obtained in (a) with the sequences in the tumor-specific proteome database of (c) and the personalized tumor proteome database of (d) to identify the MAPs; and 106 (h) identifying a candidate tumor antigen among the MAPs identified in (f), wherein a candidate tumor antigen corresponds to a MAP (1) whose sequence is not present in the personalized normal proteome database; and (2) (i) whose sequence is present in the personalized tumor proteome database; and / or (ii) whose coding sequence is overexpressed or overrepresented in said tumor k-mer database relative to said normal k-mer database.

15. The method of any of claims 1 to 14, wherein said method further comprises selecting MAPs having a length of 8 to 11 amino acids.

16. The method of any of claims 1 to 15, wherein said normal cells are thymic cells.

17. The method of claim 16, wherein said thymic cells are medullary thymic epithelial cells (mTEC).

18. The method of any of claims 1 to 17, further comprising comparing the coding sequence of said candidate tumor antigen with sequences from normal tissues.

19. The method of any of claims 1 to 18, wherein said MAPs have a length of 8 to 11 amino acids.

20. The method of any of claims 1 to 19, further comprising evaluating the binding of the candidate tumor antigen to an MHC molecule.

21. The method of claim 20, wherein said binding is evaluated by using an MHC binding prediction algorithm.

22. The method of any one of claims 1 to 21, further comprising evaluating the frequency of T cells that recognize the candidate tumor antigen in a cell population. 107 23. The method of claim 22, wherein the frequency of T cells recognizing the candidate tumor antigen is assessed by using multimeric MHC class I molecules comprising said candidate tumor antigen in their peptide-binding groove.

24. The method of any of claims 1 to 23, further comprising evaluating the ability of the candidate tumor antigen to induce T cell activation.

25. The method of claim 24, wherein the ability of the candidate tumor antigen to induce T cell activation is evaluated by measuring the cytokine production by T cells in contact with cells having said candidate tumor antigen bound to MHC class I molecules on their cell surface.

26. The method of claim 25, wherein said cytokine production comprises the production of interferon-gamma (IFN-γ).

27. The method of any of claims 1 to 26, further comprising evaluating the ability of said candidate tumor antigen to induce T cell-mediated tumor cell death and / or inhibit tumor growth.

28. A tumor antigenic peptide identified by the method defined in any of claims 1 to 27.

29. A tumor antigenic peptide comprising or consisting of one of the amino acid sequences listed in any of the SEQ ID NO: 1-39.

30. The tumor antigenic peptide according to claim 29, comprising or consisting of one of the amino acid sequences set out in any of the SEQ ID NO: 1739.

31. The tumor antigenic peptide according to claim 30, wherein said tumor antigenic peptide is a leukemia tumor antigenic peptide and comprises or consists of one of the amino acid sequences set forth in any of the SEQ ID NO: 17-28. 108 32. The tumor antigenic peptide according to claim 31, wherein said leukemia is B-cell acute lymphoblastic leukemia (B-ALL).

33. The tumor antigenic peptide according to claim 31 or 32, wherein said tumor antigenic peptide binds to a human leukocyte antigen (HLA) of the HLAA*02:01 allele and comprises or consists of one of the amino acid sequences set forth in any of SEQ ID NO: 17-19, 27 and 28.

34. The tumor antigenic peptide according to claim 31 or 32, wherein said tumor antigenic peptide binds to a human leukocyte antigen (HLA) of the HLAB*40:01 allele and comprises or consists of the amino acid sequence set out in SEQ ID NO:

20.

35. The tumor antigenic peptide according to claim 31 or 32, wherein said tumor antigenic peptide binds to a human leukocyte antigen (HLA) of the HLAA* 11:01 allele and comprises or consists of one of the amino acid sequences set forth in any of the SEQ ID NO: 21-23.

36. The tumor antigenic peptide according to claim 31 or 32, wherein said tumor antigenic peptide binds to a human leukocyte antigen (HLA) of the HLAB*08:01 allele and comprises or consists of the amino acid sequences set forth in SEQ ID NO: 24 or 25.

37. The tumor antigenic peptide according to claim 31 or 32, wherein said tumor antigenic peptide binds to a human leukocyte antigen (HLA) of the HLAB*07:02 allele and comprises or consists of the amino acid sequence set out in SEQ ID NO:

26.

38. The tumor antigenic peptide according to claim 30, wherein said tumor antigenic peptide is a lung tumor antigenic peptide and comprises or consists of one of the amino acid sequences set forth in any of the SEQ ID NO: 29-39. 109 39. The tumor antigenic peptide according to claim 38, wherein said lung tumor is a non-small cell lung cancer (NSCLC).

40. The tumor antigenic peptide according to claim 38 or 39, wherein said tumor antigenic peptide binds to a human leukocyte antigen (HLA) of the HLAA* 11:01 allele and comprises or consists of one of the amino acid sequences set forth in any of the SEQ ID NO: 29-35.

41. The tumor antigenic peptide according to claim 38 or 39, wherein said tumor antigenic peptide binds to a human leukocyte antigen (HLA) of the HLAB*07:02 allele and comprises or consists of the amino acid sequence set out in SEQ ID NO:

36.

42. The tumor antigenic peptide according to claim 38 or 39, wherein said tumor antigenic peptide binds to a human leukocyte antigen (HLA) of the HLAA*24:02 allele and comprises or consists of the amino acid sequences set forth in SEQ ID NO: 38 or 39.

43. The tumor antigenic peptide according to claim 38 or 39, wherein said tumor antigenic peptide binds to a human leukocyte antigen (HLA) of the HLAC*07:01 allele and comprises or consists of the amino acid sequence set out in SEQ ID NO:

37.

44. The tumor antigen of any of claims 29-43, which is derived from a non-protein-coding region of the genome.

45. The tumor antigen according to claim 44, wherein said non-protein-coding genome region is an intergenic region, an intronic region, a 5' untranslated region (5' UTR), a 3' untranslated region (3' UTR), or an endogenous retroelement (ERE).

46. ​​A nucleic acid encoding the tumor antigenic peptide of any of claims 28-45. 110 47. A nucleic acid according to claim 46, which is an mRNA or a viral vector.

48. A liposome comprising the tumor antigenic peptide of any of claims 28-45 or the nucleic acid according to claim 46 or 47.

49. A composition comprising the tumor antigenic peptide of any of claims 28-45, the nucleic acid according to claim 46 or 47, or the liposome according to claim 48, and a pharmaceutically acceptable carrier.

50. A vaccine comprising the tumor antigenic peptide of any of claims 28-45, the nucleic acid according to claim 46 or 47, the liposome according to claim 48 or the composition according to claim 49, and an adjuvant.

51. An isolated major histocompatibility complex (MHC) class I molecule comprising the tumor antigenic peptide of any of claims 28-45 in its peptide-binding groove.

52. The isolated MHC class I molecule according to claim 51, having the form of a multimer.

53. The isolated MHC class I molecule according to claim 52, wherein said multimer is a tetramer.

54. An isolated cell comprising the tumor antigenic peptide according to any of claims 28-45.

55. An isolated cell expressing on its surface class I major histocompatibility complex (MHC) molecules comprising the tumor antigenic peptide of any of claims 28-45 in its peptide-binding groove. cccznn / i znz / B / v 111 56. The cell according to claim 55, which is an antigen-presenting cell (APC).

57. The cell according to claim 56, wherein said APC is a dendritic cell.

58. A T cell receptor (TCR) that specifically recognizes the isolated MHC class I molecule of any of claims 51-53 and / or MHC class I molecules expressed on the cell surface of any of claims 54-57.

59. An isolated CD8+ T lymphocyte expressing the TCR on its cell surface according to claim 58.

60. A cell population comprising at least 0.5% CD8+ T lymphocytes as defined in claim 59.

61. A method for treating cancer in a subject comprising administering to the subject an effective amount of: (i) the tumor antigenic peptide of any of claims 28-45; (ii) the nucleic acid according to claim 46 or 47; (iii) the liposome according to claim 48; (iv) the composition according to claim 49; (v) the vaccine according to claim 50; (vi) the cell of any of claims 54-57; (vii) the CD8+ T lymphocytes according to claim 59; or (viii) the cell population according to claim 60.

62. The method of claim 61, wherein said cancer is leukemia.

63. The method of claim 62, wherein said leukemia is B-cell acute lymphoblastic leukemia (B-ALL).

64. The method of claim 61, wherein said cancer is lung cancer. 112 65. The method of claim 64, wherein said lung tumor is a non-small cell lung cancer (NSCLC).

66. The method of any of claims 61-65, further comprising administering at least one additional antitumor agent or therapy to the subject.

67. The method of claim 66, wherein said at least one additional antitumor agent or therapy is a chemotherapeutic agent, immunotherapy, an immune checkpoint inhibitor, radiotherapy, or surgery.

68. Use of: (i) the tumor antigenic peptide of any of claims 28-45; (ii) the nucleic acid according to claim 46 or 47; (iii) the liposome according to claim 48; (iv) the composition according to claim 49; (v) the vaccine according to claim 50; (vi) the cell of any of claims 54-57; (vii) the CD8+ T lymphocytes according to claim 59; or (viii) the cell population according to claim 60, for treating cancer in a subject.

69. Use of: (i) the tumor antigenic peptide of any of claims 28-45; (ii) the nucleic acid according to claim 46 or 47; (iii) the liposome according to claim 48; (iv) the composition according to claim 49; (v) the vaccine according to claim 50; (vi) the cell of any of claims 54-57; (vii) the CD8+ T lymphocytes according to claim 59; or (viii) the cell population according to claim 60, for the manufacture of a medicament for treating cancer in a subject.

70. Use according to claim 68 or 69, wherein said cancer is leukemia.

71. Use according to claim 70, wherein said leukemia is B-cell acute lymphoblastic leukemia (B-ALL).

72. Use according to claim 68 or 69, wherein said cancer is lung cancer. 113 Ί3. Use according to claim 72, wherein said lung tumor is non-small cell lung cancer (NSCLC).

74. The use of any of claims 68-73, further comprising the use of at least one additional antitumor agent or therapy.

75. Use according to claim 74, wherein said at least one additional antitumor agent or therapy is a chemotherapeutic agent, immunotherapy, an immune checkpoint inhibitor, radiotherapy, or surgery.