Composition and method of optimized peptide vaccine

Abstract

The present disclosure provides methods, systems, and compositions for nucleic acid and peptide sequences. The present disclosure provides nucleic acid sequences encoding at least two amino acid sequences selected from the group consisting of SEQ ID NOs: 1-6, 8-10, 12-28, and 30-41. The present disclosure also provides immunogenic peptide compositions comprising at least one peptide selected from the group consisting of SEQ ID NOs: 42-65. The present disclosure provides compositions comprising nucleic acid sequences encoding at least two amino acid sequences selected from the group consisting of SEQ ID NOs: 1550-1593. The present disclosure further provides nucleic acid sequences encoding at least one amino acid sequence selected from the group consisting of SEQ ID NOs: 1595-1661.

Description

[Technical Field]

[0001] This application claims the benefit and priority under § 119(e) of U.S. Patent Act to U.S. Patent Application No. 17 / 336,960, filed on 2 June 2021. U.S. Patent Application No. 17 / 336,960 is a continuation of U.S. Patent Application No. 17 / 100,630, filed on 20 November 2020, which is now U.S. Patent No. 11,058,751, registered on 13 July 2021. The contents of each of these documents are incorporated herein by reference in their entirety.

[0002] Furthermore, this application claims the benefit and priority under § 119(e) of the United States Patent Act to U.S. Patent Application No. 17 / 389,875, filed on 30 July 2021, which is a continuation of U.S. Patent Application No. 17 / 114,237, filed on 7 December 2020, now U.S. Patent No. 11,161,892, granted on 2 November 2021. The contents of each of these documents are incorporated herein by reference in their entirety.

[0003] Furthermore, this application claims the interests of U.S. Patent Application No. 63 / 249,235, filed on 28 September 2021, the contents of which are incorporated herein by reference in their entirety.

[0004] All patents, patent applications, and publications cited herein are incorporated herein by reference in their entirety. The disclosures of such publications are incorporated herein by reference in their entirety.

[0005] This patent disclosure contains copyrighted material. The copyright holder will not object to any facsimile copies of the patent document or patent disclosure found in the U.S. Patent and Trademark Office patent files or records, but otherwise reserves all copyrights.

[0006] Embedding by reference All documents cited herein are incorporated herein by reference in their entirety.

[0007] Sequence List This application includes a sequence listing submitted electronically in ASCII format, which is incorporated herein by reference in its entirety. This ASCII copy was created on November 15, 2021, with the filename 2215269_00123WO1_SL.txt and a size of 23,070,263 bytes.

[0008] Technical field The present invention relates, in general terms, to compositions, systems, and methods for peptide vaccines. More specifically, to compositions, systems, and methods for designing peptide vaccines for treating or preventing disease, optimized based on predicted herd immunogenicity. [Background technology]

[0009] The goal of peptide vaccines is to improve the immune response to cancer cells or pathogens by training the immune system to extend its ability to recognize and engage cells that present target peptides. Peptide vaccines can also be administered to those already affected to increase the immune response to a causally related cancer, other disease, or pathogen. Alternatively, peptide vaccines can be administered to induce the immune system to have therapeutic tolerance to one or more peptides. There is a need for peptide vaccine compositions, systems, and methods based on predictions of target peptides that will be presented to defend the host from cancer, other disease, or pathogen infection. We provide novel prophylactic and therapeutic cancer vaccines based on neoantigens introduced by RAS gene mutations occurring in numerous cancers, as well as BCR-ABL gene fusions occurring in cases of chronic myeloid leukemia (CML), acute lymphoblastic leukemia (ALL), and acute myeloid leukemia (AML), invasive ductal carcinoma, and other cancers. [Overview of the project]

[0010] In one embodiment, the present invention provides nucleic acid sequences encoding at least two amino acid sequences selected from the group consisting of SEQ ID NOs: 1, 2, 3, 4, 5, 6, 8, 9, 10, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, and 41.

[0011] In some embodiments, the nucleic acid sequence is an immunogenic composition. In some embodiments, the nucleic acid sequence is administered in vivo in an expression construct. In some embodiments, in vivo administration of the nucleic acid sequence is configured to produce at least one peptide presented by an HLA class I molecule. In some embodiments, the at least one peptide is a modified or unmodified fragment of a mutant KRAS protein. In some embodiments, the mutant KRAS protein is selected from the group consisting of KRAS G12D, KRAS G12V, KRAS G12R, KRAS G12C, and KRAS G13D. In some embodiments, the nucleic acid sequence is administered to a subject in an effective dose to prevent cancer. In some embodiments, the nucleic acid sequence is administered to a subject in an effective dose to treat cancer.

[0012] In another embodiment, the present invention provides an immunogenic peptide composition comprising at least two peptides selected from the group consisting of SEQ ID NOs: 1, 2, 3, 4, 5, 6, 8, 9, 10, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, and 41.

[0013] In some embodiments, at least one of at least two peptides is presented in the subject by an HLA class I molecule. In some embodiments, at least one of at least two peptides is a modified or unmodified fragment of a mutant KRAS protein. In some embodiments, the mutant KRAS protein is selected from the group consisting of KRAS G12D, KRAS G12V, KRAS G12R, KRAS G12C, and KRAS G13D. In some embodiments, the immunogenic peptide composition is administered to the subject in an effective amount to prevent cancer. In some embodiments, the immunogenic peptide composition is administered to the subject in an effective amount to treat cancer. In some embodiments, the immunogenic peptide composition comprises at least three peptides selected from the group consisting of SEQ ID NOs: 1, 2, 3, 4, 5, 6, 8, 9, 10, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, and 41.

[0014] In another aspect, the present invention provides a nucleic acid sequence encoding at least one amino acid sequence selected from the group consisting of SEQ ID NO: 42, SEQ ID NO: 43, SEQ ID NO: 44, SEQ ID NO: 45, SEQ ID NO: 46, SEQ ID NO: 47, SEQ ID NO: 48, SEQ ID NO: 49, SEQ ID NO: 50, SEQ ID NO: 51, SEQ ID NO: 52, SEQ ID NO: 53, SEQ ID NO: 54, SEQ ID NO: 55, SEQ ID NO: 56, SEQ ID NO: 57, SEQ ID NO: 58, SEQ ID NO: 59, SEQ ID NO: 60, SEQ ID NO: 61, SEQ ID NO: 62, SEQ ID NO: 63, SEQ ID NO: 64, and SEQ ID NO: 65.

[0015] In some embodiments, the nucleic acid sequence is an immunogenic composition. In some embodiments, the nucleic acid sequence is administered in vivo in an expression construct. In some embodiments, the in vivo administration of the nucleic acid sequence is configured to produce at least one peptide presented by HLA class II molecules. In some embodiments, the at least one peptide is a modified fragment of a mutant KRAS protein. In some embodiments, the mutant KRAS protein is selected from the group consisting of KRAS G12D, KRAS G12V, KRAS G12R, KRAS G12C, and KRAS G13D. In some embodiments, the nucleic acid sequence is administered to a subject in an effective amount to prevent cancer. In some embodiments, the nucleic acid sequence is administered to a subject in an effective amount to treat cancer.

[0016] In another aspect, the present invention provides an immunogenic peptide composition comprising at least one peptide selected from the group consisting of SEQ ID NO: 42, SEQ ID NO: 43, SEQ ID NO: 44, SEQ ID NO: 45, SEQ ID NO: 46, SEQ ID NO: 47, SEQ ID NO: 48, SEQ ID NO: 49, SEQ ID NO: 50, SEQ ID NO: 51, SEQ ID NO: 52, SEQ ID NO: 53, SEQ ID NO: 54, SEQ ID NO: 55, SEQ ID NO: 56, SEQ ID NO: 57, SEQ ID NO: 58, SEQ ID NO: 59, SEQ ID NO: 60, SEQ ID NO: 61, SEQ ID NO: 62, SEQ ID NO: 63, SEQ ID NO: 64, and SEQ ID NO: 65.

[0017] In some embodiments, at least one peptide in the immunogenic peptide composition is presented by an HLA class II molecule. In some embodiments, at least one peptide in the immunogenic peptide composition is a modified or unmodified fragment of a mutant KRAS protein. In some embodiments, the mutant KRAS protein is selected from the group consisting of KRAS G12D, KRAS G12V, KRAS G12R, KRAS G12C, and KRAS G13D. In some embodiments, the immunogenic peptide composition is administered to a subject in an effective amount to prevent cancer. In some embodiments, the immunogenic peptide composition is administered to a subject in an effective amount to treat cancer. In some embodiments, the immunogenic peptide composition comprises at least two peptides selected from the group consisting of SEQ ID NOs: 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, and 65.

[0018] In another embodiment, the present invention relates to a method for forming an immunogenic peptide composition, comprising the steps of: producing a first peptide set by selecting a plurality of unmodified peptide sequences using a processor, wherein at least one of the plurality of unmodified peptide sequences is associated with a tumor neoantigen or autoprotein; determining a plurality of peptide-HLA binding scores for each peptide sequence in the first peptide set; determining whether each peptide sequence in the first peptide set has a peptide-HLA binding score that satisfies a first threshold with respect to at least three HLA alleles; producing a second peptide set comprising the first peptide set and a plurality of modified peptide sequences, wherein each modified peptide sequence in the plurality of modified peptide sequences comprises a substitution of at least one amino acid residue in a peptide sequence in the first peptide set; and determining whether each modified peptide sequence in the second peptide set satisfies a second threshold with respect to at least three HLA alleles. The present invention provides a method comprising: a step of determining whether or not a peptide has a peptide-HLA binding score, wherein a second threshold is more restrictive than a first threshold; and a step of creating a third peptide set by selecting a subset of a second peptide set, wherein the selection includes calculating population coverage, the calculation of population coverage including excluding the peptide-HLA binding score of the modified peptide sequence with respect to the first HLA allele if the peptide-HLA binding score of the unmodified peptide sequence associated with the modified peptide sequence does not meet the first threshold with respect to the first HLA allele, and the selected subset has population coverage exceeding the third threshold; performing an experimental assay to obtain a peptide-HLA immunogenicity metric for at least one peptide sequence in the third peptide set; and forming an immunogenic peptide composition comprising at least one peptide sequence in the third peptide set on which the experimental assay was performed.

[0019] In some embodiments, selecting multiple unmodified peptide sequences to create a first peptide set involves sliding a window of size n across amino acid sequences encoding tumor neoantigens or autoproteins, where n is approximately 8 to 25 amino acid lengths, and n is the length of each peptide sequence in the multiple unmodified peptide sequences in the first peptide set. In some embodiments, each peptide sequence in the first peptide set binds to an HLA class I molecule or an HLA class II molecule. In some embodiments, the method further includes filtering the first peptide set to exclude peptide sequences having a predictive binding core that includes the target residue at the anchor position. In some embodiments, the method further includes substituting at least one amino acid residue in each peptide sequence in the first peptide set, where at least one amino acid residue is present at the anchor position for at least one peptide sequence in the first peptide set. In some embodiments, a first threshold is a binding affinity of less than approximately 1000 nM. In some embodiments, a second threshold is a binding affinity of less than approximately 500 nM. In some embodiments, population coverage is calculated based on the frequency of HLA haplotypes in the human population. In some embodiments, population coverage is calculated based on the frequency of at least three HLA alleles in the human population. In some embodiments, multiple unmodified peptide sequences are derived from tumor neoantigens or autoproteins present in the subject. In some embodiments, a third threshold is a proportion of the human population of approximately 0.7 to approximately 0.8. In some embodiments, the tumor neoantigen or autoprotein is associated with cancer, and the cancer is selected from the group consisting of pancreatic, colon, rectal, kidney, bronchus, lung, uterus, cervix, bladder, liver, and stomach.

[0020] In another aspect, the present invention relates to a method for forming an immunogenic peptide composition, comprising the steps of: producing a first peptide set by selecting a plurality of unmodified peptide sequences using a processor, wherein the plurality of unmodified peptide sequences are associated with tumor neoantigens or autoproteins; determining a plurality of peptide-HLA immunogenicity metrics for each peptide sequence in the first peptide set; determining whether each peptide sequence in the first peptide set has a peptide-HLA immunogenicity metric that satisfies a first threshold with respect to at least three HLA alleles; producing a second peptide set comprising the first peptide set and a plurality of modified peptide sequences, wherein each modified peptide sequence in the plurality of modified peptide sequences comprises a substitution of at least one amino acid residue in a peptide sequence in the first peptide set; and determining whether each modified peptide sequence in the second peptide set has a peptide-HLA immunogenicity metric that satisfies a second threshold with respect to at least three HLA alleles. The present invention provides a method comprising: a step of determining whether or not a second threshold is more restrictive than a first threshold; and a step of creating a third peptide set by selecting a subset of a second peptide set, the selection comprising calculating predicted vaccine performance, the calculation of predicted vaccine performance comprising excluding the peptide-HLA immunogenicity metric of the modified peptide sequence with respect to the first HLA allele if the peptide-HLA immunogenicity metric of the unmodified peptide sequence associated with the modified peptide sequence does not satisfy the first threshold with respect to the first HLA allele, and the predicted vaccine performance is a function of the unexcluded peptide-HLA immunogenicity metric of each peptide sequence in the third peptide set; performing an experimental assay to obtain a peptide-HLA immunogenicity metric for at least one peptide sequence in the third peptide set; and forming an immunogenic peptide composition comprising at least one peptide sequence in the third peptide set on which the experimental assay has been performed.

[0021] In some embodiments, selecting multiple unmodified peptide sequences to create a first peptide set involves sliding a window of size n across amino acid sequences encoding tumor neoantigens or autoproteins, where n is approximately 8 to 25 amino acid lengths, and n is the length of each peptide sequence in the multiple unmodified peptide sequences in the first peptide set. In some embodiments, each peptide sequence in the first peptide set binds to an HLA class I molecule or an HLA class II molecule. In some embodiments, the method further includes filtering the first peptide set to exclude peptide sequences having a predictive binding core that includes the target residue at the anchor position. In some embodiments, the method further includes substituting at least one amino acid residue in each peptide sequence in the first peptide set, where at least one amino acid residue is present at the anchor position for at least one peptide sequence in the first peptide set. In some embodiments, the first threshold is a binding affinity of less than approximately 1000 nM.

[0022] In another aspect, the present invention relates to a method for forming an immunogenic peptide composition, comprising the steps of: producing a first peptide set by selecting a plurality of unmodified peptide sequences using a processor, wherein each peptide sequence of the plurality of unmodified peptide sequences is associated with a tumor neoantigen or autoprotein; determining a plurality of peptide-HLA binding scores for each peptide sequence in the first peptide set; determining whether each peptide sequence in the first peptide set has a peptide-HLA binding score that satisfies a first threshold with respect to at least three HLA alleles; producing a second peptide set comprising the first peptide set and a plurality of modified peptide sequences, wherein each modified peptide sequence of the plurality of modified peptide sequences comprises a substitution of at least one amino acid residue in a peptide sequence in the first peptide set; and determining whether each modified peptide sequence in the second peptide set has a peptide-HLA binding score that satisfies a second threshold with respect to at least three HLA alleles. The present invention provides a method comprising: a step of determining whether or not a second threshold is more restrictive than a first threshold; a step of creating a third peptide set by selecting a subset of a second peptide set, the selection comprising calculating predicted vaccine performance, the calculation of predicted vaccine performance comprising excluding the peptide-HLA binding score of a modified peptide sequence with respect to a first HLA allele if the peptide-HLA binding score of an unmodified peptide sequence associated with a modified peptide sequence does not meet a first threshold with respect to a first HLA allele, and the predicted vaccine performance is a function of the unexcluded peptide-HLA binding scores of each peptide sequence in the third peptide set; performing an experimental assay to obtain a peptide-HLA immunogenicity metric for at least one peptide sequence in the third peptide set; and forming an immunogenic peptide composition comprising at least one peptide sequence in the third peptide set on which the experimental assay has been performed.

[0023] In some embodiments, the second threshold is based on data obtained from one or more experimental assays. In some embodiments, the performance of the predicted vaccine is further a function of the peptide-HLA immunogenicity metric of at least one modified peptide sequence among a plurality of modified peptide sequences bound to the second HLA allele among a plurality of HLA alleles, where the first peptide sequence in the first peptide set is predicted to bind to the second HLA allele of at least three HLA alleles by a first binding core, the first binding core is the binding core of the first peptide sequence, the first binding core is identical to the second binding core, the first and second binding cores contain amino acid positions within the peptide sequence, and the second binding core is the binding core of at least one modified peptide sequence among a plurality of modified peptide sequences bound to the second HLA allele.

[0024] In another aspect, the present invention relates to a method for forming an immunogenic peptide composition, comprising the steps of: producing a first peptide set by selecting a plurality of unmodified peptide sequences using a processor, wherein at least one of the plurality of unmodified peptide sequences is associated with a tumor neoantigen or autoprotein; determining a plurality of peptide-HLA binding scores for each peptide sequence in the first peptide set; determining whether each peptide sequence in the first peptide set has a peptide-HLA binding score that satisfies a first threshold with respect to at least three HLA alleles; producing a second peptide set comprising the first peptide set and a plurality of modified peptide sequences, wherein each modified peptide sequence in the plurality of modified peptide sequences comprises a substitution of at least one amino acid residue in a peptide sequence in the first peptide set; and determining whether each modified peptide sequence in the second peptide set satisfies a first threshold with respect to at least three HLA alleles. The present invention provides a method comprising: a step of determining whether a peptide has a peptide-HLA binding score that satisfies a second threshold, wherein the second threshold is more restrictive than a first threshold; a step of creating a third peptide set by selecting a subset of a second peptide set, wherein the selection includes calculating predicted vaccine performance based on the target HLA type, and the calculation of predicted vaccine performance includes excluding the peptide-HLA binding score of the modified peptide sequence with respect to the first HLA allele if the peptide-HLA binding score of the unmodified peptide sequence associated with the modified peptide sequence does not satisfy the first threshold with respect to the first HLA allele; a step of performing an experimental assay to obtain a peptide-HLA immunogenicity metric for at least one peptide sequence in the third peptide set; and a step of forming an immunogenic peptide composition comprising at least one peptide sequence in the third peptide set on which the experimental assay has been performed.

[0025] In some embodiments, each peptide sequence in the first peptide set binds to an HLA class I molecule or an HLA class II molecule. In some embodiments, multiple unmodified peptide sequences are derived from tumor neoantigens or autoproteins present in the subject. In some embodiments, at least three HLA alleles are present in the subject's HLA type. In some embodiments, multiple unmodified peptide sequences are derived from tumor neoantigens or autoproteins present in the subject. In some embodiments, multiple unmodified peptide sequences are derived from tumor neoantigens or autoproteins present in the subject. In some embodiments, the immunogenic peptide composition includes a nucleic acid sequence encoding the amino acid sequence of at least one peptide sequence in the third peptide set. In some embodiments, the immunogenic peptide composition includes a nucleic acid sequence encoding the amino acid sequence of at least one peptide sequence in the third peptide set.

[0026] In another aspect, the present invention provides a composition comprising nucleic acid sequences encoding at least two amino acid sequences selected from the group consisting of SEQ ID NOs: 1550 to 1593.

[0027] In some embodiments, the composition is immunogenic. In some embodiments, the nucleic acid sequence is administered in vivo in a construct for expression. In some embodiments, in vivo administration of the nucleic acid sequence is configured to produce at least one peptide presented by an HLA class I molecule. In some embodiments, the at least one peptide is a modified or unmodified fragment of a BCL-ABL gene fusion. In some embodiments, the BCR-ABL gene fusion is b3a2 or b2a2. In some embodiments, the nucleic acid sequence is administered to a subject in an effective dose to prevent cancer. In some embodiments, the nucleic acid sequence is administered to a subject in an effective dose to treat cancer.

[0028] In another embodiment, the present invention provides a composition comprising at least two aminopeptides selected from the group consisting of SEQ ID NOs: 1550 to 1593.

[0029] In some embodiments, at least one of at least two peptides is presented by an HLA class I molecule in the subject. In some embodiments, at least one peptide in the immunogenic peptide composition is a modified or unmodified fragment of a BCL-ABL gene fusion. In some embodiments, the BCR-ABL gene fusion is b3a2 or b2a2. In some embodiments, the immunogenic peptide composition is administered to the subject in an effective amount to prevent cancer. In some embodiments, the immunogenic peptide composition is administered to the subject in an effective amount to treat cancer. In another embodiment, the present invention provides a composition comprising at least two peptides selected from the group consisting of SEQ ID NOs: 1550-1593.

[0030] In another aspect, the present invention provides a nucleic acid sequence encoding at least one amino acid sequence selected from the group consisting of SEQ ID NOs: 1595 to 1661.

[0031] In some embodiments, the composition is immunogenic. In some embodiments, the nucleic acid sequence is administered in vivo in a construct for expression. In some embodiments, in vivo administration of the nucleic acid sequence is configured to produce at least one peptide presented by an HLA class II molecule. In some embodiments, at least one amino acid sequence is derived from a modified fragment of a BCL-ABL gene fusion. In some embodiments, the BCR-ABL gene fusion is b3a2 or b2a2. In some embodiments, the nucleic acid sequence is administered to a subject in an effective dose to prevent cancer. In some embodiments, the nucleic acid sequence is administered to a subject in an effective dose to treat cancer.

[0032] In another aspect, the present invention provides a nucleic acid sequence encoding at least one amino acid sequence selected from the group consisting of SEQ ID NOs: 1595 to 1661.

[0033] In some embodiments, at least one peptide is presented by an HLA class II molecule in the subject. In some embodiments, at least one peptide in the immunogenic peptide composition is a modified or unmodified fragment of a BCL-ABL gene fusion. In some embodiments, the BCR-ABL gene fusion is b3a2 or b2a2. In some embodiments, the immunogenic peptide composition is administered to the subject in an effective dose to prevent cancer. In some embodiments, the immunogenic peptide composition is administered to the subject in an effective dose to treat cancer. In some embodiments, the immunogenic peptide composition comprises at least two peptides selected from the group consisting of SEQ ID NOs: 1595-1661.

[0034] In another aspect, the present invention relates to a method for forming an immunogenic peptide composition, comprising the steps of: producing a first peptide set by selecting a plurality of unmodified peptide sequences using a processor, wherein at least one of the plurality of unmodified peptide sequences is associated with a tumor neoantigen, a pathogen proteome, or an autoprotein; determining a plurality of peptide-HLA binding scores for each peptide sequence in the first peptide set; determining whether each peptide sequence in the first peptide set has a peptide-HLA binding score that satisfies a first threshold with respect to at least three HLA alleles; and producing a second peptide set comprising the first peptide set and a plurality of modified peptide sequences, wherein each modified peptide sequence in the plurality of modified peptide sequences comprises a substitution of at least one amino acid residue in a peptide sequence in the first peptide set; and each of the second peptide sets The present invention provides a method comprising the steps of: determining multiple peptide-HLA binding scores for each peptide sequence; creating a third peptide set by selecting a subset of a second peptide set, wherein the selection includes calculating population coverage, and the calculation of population coverage includes excluding the peptide-HLA binding score for the first HLA allele of the modified peptide sequence if the peptide-HLA binding score of the unmodified peptide sequence associated with the modified peptide sequence does not meet a first threshold with respect to the first HLA allele, and the selected subset has population coverage exceeding the third threshold; performing an experimental assay to obtain a peptide-HLA immunogenicity metric for at least one peptide sequence in the third peptide set; and forming an immunogenic peptide composition containing at least one peptide sequence from the third peptide set on which the experimental assay was performed.

[0035] In some embodiments, selecting multiple unmodified peptide sequences to create a first peptide set involves sliding a window of size n over at least a portion of amino acid sequences encoding tumor neoantigens, pathogen proteomes, or autoproteins, where n is approximately 8 to 25 amino acid lengths, and n is the length of each peptide sequence in the multiple unmodified peptide sequences of the first peptide set. In some embodiments, each peptide sequence in the first peptide set binds to an HLA class I molecule or an HLA class II molecule. In some embodiments, the first peptide set is further filtered to exclude peptide sequences having a predictive binding core that includes the target amino acid residue at the anchor position. In some embodiments, at least one amino acid residue is substituted in each peptide sequence of the first peptide set, so that for at least one peptide sequence in the first peptide set, at least one amino acid residue is present at the anchor position. In some embodiments, the first threshold is a binding affinity of less than approximately 1000 nM. In some embodiments, population coverage is calculated with respect to at least three HLA alleles. In some embodiments, population coverage is calculated based on the frequency of HLA haplotypes in the human population. In some embodiments, population coverage is calculated based on the frequency of at least three HLA alleles in the human population. In some embodiments, the multiple unmodified peptide sequences are derived from tumor neoantigens, pathogen proteomes, or autoproteins present in the subject. In some embodiments, the second threshold is a proportion of the human population of about 0.7 to about 0.8. In some embodiments, the tumor neoantigen or autoprotein is associated with cancer, which is selected from the group consisting of pancreatic, colon, rectal, kidney, bronchus, lung, uterus, cervix, bladder, liver, and stomach. In some embodiments, the pathogen proteome is associated with pathogen infection in the human subject. In some embodiments, the immunogenic peptide composition includes a nucleic acid sequence encoding the amino acid sequence of at least one peptide sequence of a third peptide set.

[0036] In another aspect, the present invention relates to a method for forming an immunogenic peptide composition, comprising the steps of: producing a first peptide set by selecting a plurality of unmodified peptide sequences using a processor, wherein the plurality of unmodified peptide sequences are associated with tumor neoantigens, pathogen proteomes, or autoproteins; determining a plurality of peptide-HLA immunogenicity metrics for each peptide sequence in the first peptide set; determining whether each peptide sequence in the first peptide set has a peptide-HLA immunogenicity metric that satisfies a threshold with respect to at least three HLA alleles; producing a second peptide set comprising the first peptide set and a plurality of modified peptide sequences, wherein each modified peptide sequence in the plurality of modified peptide sequences comprises a substitution of at least one amino acid residue in a peptide sequence in the first peptide set; and determining whether each peptide sequence in the second peptide set has a plurality of peptide-HLA immunogenicity metrics that satisfy a threshold with respect to at least three HLA alleles. The present invention provides a method comprising the steps of: determining a trick; creating a third peptide set by selecting a subset of a second peptide set, wherein the selection includes calculating predicted vaccine performance, the calculation of predicted vaccine performance including excluding the peptide-HLA immunogenicity metric of the modified peptide sequence with respect to the first HLA allele if the peptide-HLA immunogenicity metric of the unmodified peptide sequence associated with the modified peptide sequence does not meet a threshold with respect to the first HLA allele, and the predicted vaccine performance is a function of the peptide-HLA immunogenicity metric of each peptide sequence in the third peptide set; performing an experimental assay to obtain a peptide-HLA immunogenicity metric for at least one peptide sequence of the third peptide set; and forming an immunogenic peptide composition comprising at least one peptide sequence of the third peptide set on which the experimental assay has been performed.

[0037] In some embodiments, selecting multiple unmodified peptide sequences to create a first peptide set involves sliding a window of size n over at least a portion of amino acid sequences encoding tumor neoantigens, pathogen proteomes, or autoproteins, where n is approximately 8 to 25 amino acid lengths, and n is the length of each peptide sequence in the multiple unmodified peptide sequences of the first peptide set. In some embodiments, each peptide sequence in the first peptide set binds to an HLA class I molecule or an HLA class II molecule. In some embodiments, the first peptide set is further filtered to exclude peptide sequences having a predictive binding core containing the target amino acid residue at the anchor position. In some embodiments, at least one amino acid residue is substituted in each peptide sequence of the first peptide set. In some embodiments, the threshold is a binding affinity of less than approximately 1000 nM. In some embodiments, at least three HLA alleles are present in the target HLA type. In some embodiments, the multiple peptide sequences are derived from tumor neoantigens, pathogen proteomes, or autoproteins present in the target. In some embodiments, the immunogenic peptide composition includes a nucleic acid sequence encoding the amino acid sequence of at least one peptide sequence of a third peptide set.

[0038] In another aspect, the present invention relates to a method for forming an immunogenic peptide composition, comprising the steps of: producing a first peptide set by selecting a plurality of unmodified peptide sequences using a processor, wherein each peptide sequence of the plurality of unmodified peptide sequences is associated with a tumor neoantigen, a pathogen proteome, or an autoprotein; determining a plurality of peptide-HLA binding scores for each peptide sequence in the first peptide set; determining whether each peptide sequence in the first peptide set has a peptide-HLA binding score that satisfies a threshold with respect to at least three HLA alleles; producing a second peptide set comprising the first peptide set and a plurality of modified peptide sequences, wherein each modified peptide sequence of the plurality of modified peptide sequences comprises a substitution of at least one amino acid residue in a peptide sequence in the first peptide set; and determining whether each peptide sequence in the second peptide set has a plurality of peptide- The present invention provides a method comprising the steps of: determining an HLA binding score; creating a third peptide set by selecting a subset of a second peptide set, wherein the selection includes calculating predicted vaccine performance, the calculation of predicted vaccine performance including excluding the peptide-HLA binding score of the modified peptide sequence with respect to the first HLA allele if the peptide-HLA binding score of the unmodified peptide sequence associated with the modified peptide sequence does not meet a threshold for the first HLA allele, and the predicted vaccine performance is a function of the peptide-HLA binding score of each peptide sequence in the third peptide set; performing an experimental assay to obtain a peptide-HLA immunogenicity metric for at least one peptide sequence of the third peptide set; and forming an immunogenic peptide composition comprising at least one peptide sequence of the third peptide set on which the experimental assay was performed.

[0039] In some embodiments, a second threshold is determined from data obtained from one or more experimental assays. In some embodiments, the performance of the predicted vaccine is further a function of the peptide-HLA immunogenicity metric of at least one modified peptide sequence of the second peptide set with respect to the second HLA allele, where the first peptide sequence of the first peptide set is predicted to bind to the second HLA allele by a first binding core, the first binding core is the binding core of the first peptide sequence, the first binding core is identical to the second binding core, and the first and second binding cores each contain an amino acid position within the peptide sequence, the second binding core is the binding core of at least one modified peptide sequence. In some embodiments, the multiple peptide sequences are derived from tumor neoantigens, pathogen proteomes, or autoproteins present in the subject.

[0040] In another aspect, the present invention relates to a method for forming an immunogenic peptide composition, comprising the steps of: producing a first peptide set by selecting a plurality of unmodified peptide sequences using a processor, wherein at least one of the plurality of unmodified peptide sequences is associated with a tumor neoantigen, a pathogen proteome, or an autoprotein; determining a plurality of peptide-HLA binding scores for each peptide sequence in the first peptide set; determining whether each peptide sequence in the first peptide set has a peptide-HLA binding score that satisfies a threshold with respect to at least three HLA alleles; producing a second peptide set comprising the first peptide set and a plurality of modified peptide sequences, wherein each modified peptide sequence in the plurality of modified peptide sequences comprises a substitution of at least one amino acid residue in a peptide sequence in the first peptide set; and producing a second peptide set comprising the first peptide set and a plurality of modified peptide sequences. The present invention provides a method comprising the steps of: determining multiple peptide-HLA binding scores for each peptide sequence in a peptide set; creating a third peptide set by selecting a subset of a second peptide set, wherein the selection includes calculating predicted vaccine performance based on the target HLA type, and the calculation of predicted vaccine performance includes excluding the peptide-HLA binding score of the modified peptide sequence for the first HLA allele if the peptide-HLA binding score of the unmodified peptide sequence associated with the modified peptide sequence does not meet a threshold with respect to the first HLA allele; performing an experimental assay to obtain a peptide-HLA immunogenicity metric for at least one peptide sequence of the third peptide set; and forming an immunogenic peptide composition containing at least one peptide sequence of the third peptide set on which the experimental assay was performed.

[0041] In some embodiments, each peptide sequence in the first peptide set binds to an HLA class I molecule or an HLA class II molecule. In some embodiments, multiple peptide sequences are derived from tumor neoantigens, pathogen proteomes, or autoproteins present in the subject.

[0042] The following figures illustrate exemplary embodiments of the present invention. [Brief explanation of the drawing]

[0043] [Figure 1] Figure 1 is a flowchart of the vaccine optimization method. [Figure 2] Figure 2 is a flowchart of the vaccine optimization method using seed set compression. [Figure 3] Figure 3 is a graph showing the predicted population coverage of MHC class I vaccines by vaccine size for KRAS G12D, KRAS G12V, KRAS G12R, KRAS G12C, and KRAS G13D targets. [Figure 4] Figure 4 is a graph showing the predicted population coverage of MHC class II vaccines by vaccine size for KRAS G12D, KRAS G12V, KRAS G12R, KRAS G12C, and KRAS G13D targets. [Figure 5] Figure 5 is a graph showing the predicted population coverage by vaccine size for MHC class I vaccines containing heterocritical peptides for BCR-ABL b3a2 fusions, at least one peptide-HLA hit (circles), at least three peptide-HLA hits (triangles), and at least five peptide-HLA hits (squares). The dashed lines show the predicted population coverage for BCR-ABL b3a2 fusion vaccines without heterocritical peptides for at least one (upper dashed line) and five (lower dashed line) peptide-HLA hits per individual. [Figure 6]Figure 6 is a graph showing the predicted population coverage by vaccine size for MHC class II vaccines containing heterocritical peptides for BCR-ABL b3a2 fusions, at least one peptide-HLA hit (circles), at least three peptide-HLA hits (triangles), and at least five peptide-HLA hits (squares). The dashed line shows the predicted population coverage for BCR-ABL b3a2 fusion vaccines without heterocritical peptides for at least one peptide-HLA hit per individual. [Figure 7] Figure 7 is a graph showing the predicted population coverage by vaccine size for MHC class I vaccines containing heterocritical peptides for BCR-ABL b2a2 fusions, at least one peptide-HLA hit (circles), at least three peptide-HLA hits (triangles), and at least five peptide-HLA hits (squares). The dashed line shows the predicted population coverage for BCR-ABL b3a2 fusion vaccines without heterocritical peptides for at least one peptide-HLA hit per individual. [Figure 8] Figure 8 is a graph showing the predicted population coverage by vaccine size for MHC class II vaccines containing heterocritical peptides for BCR-ABL b2a2 fusions, at least one peptide-HLA hit (circles), at least three peptide-HLA hits (triangles), and at least five peptide-HLA hits (squares). The dashed lines show the predicted population coverage for BCR-ABL b3a2 fusion vaccines without heterocritical peptides for at least one (upper dashed line) and five (lower dashed line) peptide-HLA hits per individual. [Figure 9] Figure 9 shows the predicted peptide-HLA hits for KRAS G12V vaccine by vaccine size for HLA diplotypes HLA-A02:03, HLA-A11:01, HLA-B55:02, HLA-B58:01, HLA-C03:02, and HLA-C03:03. [Figure 10]Figure 10 shows the probabilities of disease presentation in the pancreas, colon / rectum, and bronchi / lung, as well as the corresponding probabilities of target presentation for various mutant protein targets. [Figure 11] Figure 11 is a flowchart of the optimization method for multiple-target (combination) vaccines. [Figure 12] Figure 12 shows the predicted population coverage by vaccine size for pancreatic cancer multi-target (combination) MHC class I vaccines targeting KRAS G12D, KRAS G12V, and KRAS G12R. The dashed line shows the predicted population coverage for pancreatic cancer combination vaccines that do not contain MHC class I heterocritical peptides. [Figure 13] Figure 13 shows the predicted population coverage by vaccine size for pancreatic cancer multi-target (combination) MHC class II vaccines targeting KRAS G12D, KRAS G12V, and KRAS G12R. The dashed line shows the predicted population coverage for pancreatic cancer combination vaccines that do not contain MHC class II heterocritical peptides. [Figure 14] This script demonstrates a Python implementation example of the MergeMulti function for the combination vaccine design procedure. [Modes for carrying out the invention]

[0044] In some embodiments, the disclosure provides peptide vaccines incorporating peptide sequences that will be presented by major histocompatibility complex (MHC) molecules on cells to train the immune system to recognize cancer or pathogen-affected cells. In some embodiments, the disclosure provides peptide vaccines incorporating peptide sequences that will be presented by major histocompatibility complex (MHC) molecules on cells to induce therapeutic tolerance in antigen-specific immunotherapy for autoimmune diseases (Alhadj Ali et al., 2017, Gibson, et al. 2015). In some embodiments, the peptide vaccine is a composition comprising one or more peptides. In some embodiments, the peptide vaccine is an mRNA or DNA construct that encodes one or more peptides and is administered in vivo for expression.

[0045] Peptide presentation by MHC molecules is necessary, but insufficient, for peptides to be immunogenic and for the resulting peptide-MHC complex to be recognized by the organism's T cells to induce T cell activation, proliferation, and immunomemory. In some embodiments, peptide presentation by MHC molecules (e.g., HLA alleles) (e.g., peptide immunogenicity requiring peptide binding) is scored (e.g., measured as a peptide-HLA binding score) using ELISPOT (Slota et al., 2011) or multiplexing of antigen-specific T cell receptors using a combination of immunoassay and immunoreceptor sequencing (MIRA) assays (Klinger et al., 2015). In some embodiments, experimental data from assays such as ELISPOT (Slota et al., 2011) or multiplexing of antigen-specific T cell receptors using a combination of immunoassay and immunoreceptor sequencing (MIRA) assays (Klinger et al., 2015) can be used to generate peptide-HLA immunogenicity metrics for peptides and HLA alleles in a given experimental setting or organism. In some embodiments, experimental data from assays such as ELISPOT (Slota et al., 2011) or multiplexing of antigen-specific T cell receptors using a combination of immunoassay and immunoreceptor sequencing (MIRA) assays (Klinger et al., 2015) can be combined with machine learning-based predictions to score peptide presentation (e.g., binding affinity) by MHC molecules (e.g., HLA alleles) (e.g., measured as a peptide-HLA binding score) or to determine peptide-immunogenicity metrics. In some embodiments, MHCflurry or NetMHCpan (Reynisson et al., 2020) calculation methods (known in the art) are used to predict MHC class I presentation of peptides by HLA alleles (see Table 1). In some embodiments, NetMHCIIpan calculation methods (Reynisson et al., 2020) are used to predict MHC class II presentation of peptides by HLA alleles (see Table 2).

[0046] In some embodiments, MHCflurry (Odonnell et al., 2018, Odonnell et al., 2020; these documents are incorporated herein by reference in their entirety), NetMHCpan (Reynisson et al., 2020; this document is incorporated herein by reference in its entirety), and NetMHCIIpan (Reynisson et al., 2020) are used to predict either MHC class I (MHCflurry, NetMHCpan) or class II (NetMHCIIpan) presentation of a peptide by an HLA allele. In other embodiments, other methods for determining peptide-HLA binding are used, such as those disclosed in International Publication No. 2005 / 042698, which is incorporated herein by reference in its entirety. NetMHCpan-4.1 and NetMHCIIpan-4.0 use the NNAlign_MA algorithm (Alvarez et al., 2019; this entire document is incorporated herein by reference) to predict peptide-HLA binding. NNAlign_MA is then based on the NNAlign (Nielsen et al., 2009, Nielsen et al., 2017; this entire document is incorporated herein by reference) neural network. NetMHCpan-4.1 (Reynisson et al., 2020) uses an NNAlign_MA network with at least 180 inputs (9 × 20 = 180 inputs) describing the peptide sequence. Networks with both 56 and 66 hidden neurons as well as two outputs are used (Alvarez et al., 2019). Each network architecture (56 or 66 hidden neurons) is trained with five different random parameter initializations and five-fold cross-validation, resulting in a total of 50 individual trained networks (2 architectures × 5 initializations × 5 cross-validations).These 50 trained networks are used as an ensemble consisting of 25 networks with at least 10,800 parameters (180 inputs × 56 neurons) and 25 networks with at least 11,880 parameters (180 inputs × 66 neurons). Therefore, the ensemble of 50 networks in NetMHCpan-4.1 consists of at least 567,000 parameters that must be evaluated in at least 567,000 arithmetic operations to compute peptide-MHC bindings. NetMHCpan-4.1 (Reynisson et al., 2020) uses an NNAlign_MA network with at least 180 inputs (9 × 20 = 180 inputs) describing the peptide sequence. Networks with 2, 10, 20, 40, and 60 hidden neurons and 2 outputs are used (Alvarez et al., 2019). Each network architecture (2, 10, 20, 40, and 60 hidden neurons) is trained with 10 different random parameter initializations and quintuple cross-validation, resulting in a total of 250 individual trained networks (5 architectures × 10 initializations × 5 cross-validations). These 250 trained networks are used as ensembles, with 50 networks having at least 360 parameters (180 inputs × 2 neurons), 50 networks having at least 1800 parameters (180 inputs × 10 neurons), 50 networks having at least 3600 parameters (180 inputs × 20 neurons), 50 networks having at least 7200 parameters (180 inputs × 40 neurons), and 50 networks having at least 10,800 parameters (180 inputs × 60 neurons). Therefore, the ensemble of 250 networks in NetMHCpan-4.1 consists of at least 1,188,000 parameters that must be evaluated with at least 1,188,000 arithmetic operations to calculate peptide-MHC bonds.

[0047] Peptides, when bound to the grooves of MHC molecules, are presented by MHC molecules, transported to the cell surface, and can be recognized by T cell receptors. Target peptides refer to exogenous peptides or self-peptides. In some embodiments, peptides that are part of the normal proteome in a healthy individual are self-peptides, and peptides that are not part of the normal proteome are exogenous peptides. In some embodiments, target peptides may be part of the normal proteome exhibiting abnormal expression (e.g., cancer-testis antigens such as NY-ESO-1). Exogenous peptides may be generated by mutations in normal self-proteins within tumor cells that produce epitopes called neoantigens, or by pathogen infection. In some embodiments, neoantigens are any partial sequences of human proteins, and the partial sequence contains one or more altered amino acids or protein modifications that do not appear in healthy individuals. Therefore, in this disclosure, exogenous peptide refers to an amino acid sequence encoding a fragment of a target protein / peptide (or full-length protein / peptide), and the target protein / peptide consists of a neoantigen protein, pathogen proteome, or any other undesirable protein that is non-self and is expected to be bound and presented by an HLA allele.

[0048] For example, KRAS gene mutations are the most frequently mutated oncogenes in cancer, yet they are extremely difficult to treat with small molecule therapies. The KRAS protein is part of a signaling pathway that controls cell proliferation, and point mutations in this protein can lead to constitutive pathway activation and uncontrolled cell proliferation. Single-amino acid KRAS mutations result in only slight changes to the protein structure, making it difficult to develop small molecule drugs that recognize mutant-specific binding pockets and inactivate KRAS signaling. KRAS oncogenic mutations include mutations at position 12 to glycine-to-aspartic acid (G12D), glycine-to-valine (G12V), glycine-to-arginine (G12R), or glycine-to-cystine (G12C); or mutations at position 13 to glycine-to-aspartic acid (G13D). Corresponding exogenous peptides include these mutations. KRAS is a member of the RAS gene family, which also includes HRAS and NRAS. KRAS, HRAS, and NRAS have identical sequences from residue 1 to residue 86. Therefore, all vaccine and peptide sequences described herein for mutations in one RAS family member can be used for the same mutation in any other RAS family member (for example, the KRAS G12D vaccine is also a vaccine for HRAS G12D).

[0049] BCR-ABL mutations result from abnormal junctioning between the BCR gene on chromosome 22 and the ABL gene on chromosome 9, leading to a fusion of the two genes on chromosome 22. Differences in the fusion product formed result in different BCR-ABL transcripts, with b3a2 (also known as e14a2) and b2a2 (also known as e13a2) being the most common. In a study of 200 patients with BCR-ABL, 42% expressed b2a2, 41% expressed b3a2, and 18% expressed both transcripts (Jain et al., 2016). Abnormal b2a2 and b3a2 BCR-ABL fusions create novel protein sequences containing exogenous peptides at the BCL-ABL junction. This specification discloses methods for using such exogenous peptides and their derivatives as neoantigen epitopes for vaccine design.

[0050] A challenge in peptide vaccine design is the diversity of human MHC alleles (HLA alleles), each exhibiting a specific priority for the peptide sequences they present. Human leukocyte antigen (HLA) loci located within the MHC gene encode HLA class I and class II molecules. There are three classical class I loci (HLA-A, HLA-B, and HLA-C) and three loci encoding class II molecules (HLA-DR, HLA-DQ, and HLA-DP). An individual's HLA type describes the alleles they possess at each of these loci. Peptides approximately 8–11 residues in length can bind to HLA class I (or MHC class I) molecules, while peptides approximately 13–25 residues in length bind to HLA class II (or MHC class II) molecules (Rist et al., 2013; Chicz et al., 1992). Human populations originating from different geographical regions exhibit different frequencies of HLA alleles, resulting in linkage disequilibrium between HLA loci and thus population-specific haplotype frequencies. In some embodiments, methods for producing effective vaccines are disclosed, which include taking into account HLA allele frequencies and linkage disequilibrium between HLA genes in a target population in order to achieve a set of peptides that are likely to be robustly presented.

[0051] This disclosure provides compositions, systems, and methods for designing vaccines that generate immunity against one or more targets. In some embodiments, the target is a neoantigen protein sequence, a pathogen proteome, or any other undesirable protein sequence that is non-self and is expected to be bound and presented by an HLA molecule (also referred herein as an HLA allele). When a target is present in an individual, this may result in multiple peptide sequences presented by various HLA alleles. In some embodiments, it may be desirable to create a vaccine containing selected self-peptides, and such selected self-peptides are therefore considered target peptides for this purpose.

[0052] The term peptide-HLA binding is defined as the binding of a peptide to an HLA allele, and can be predicted computationally, observed experimentally, or predicted computationally using experimental observation. Metrics for peptide-HLA binding can be expressed as affinity, percentile rank, binary at a given threshold, probability, or other metrics known in the art. The term peptide-HLA immunogenicity metric is defined as the activation of T cells based on the recognition of a peptide upon HLA allele binding. The term peptide-HLA immunogenicity score is another term for the peptide-HLA immunogenicity metric, and these terms are synonymous. Peptide-HLA immunogenicity metrics can vary between individuals, and metrics for peptide-HLA immunogenicity can be expressed as probability, binary index, or other metrics relating to the likelihood that a peptide-HLA combination will be immunogenic. In some embodiments, peptide-HLA immunogenicity is defined as the induction of immune tolerance based on the recognition of a peptide upon HLA allele binding. Peptide-HLA immunogenicity metrics can be computationally predicted, experimentally observed, or computationally predicted using experimental observation. In some embodiments, peptide-HLA immunogenicity requires peptide-HLA binding, so the peptide-HLA immunogenicity metric is based solely on peptide-HLA binding. In some embodiments, peptide-HLA immunogenicity data or computational predictions of peptide-HLA immunogenicity can be combined with peptide presentation scores in the methods disclosed herein. One method of combining scores is to attribute a peptide to the HLA allele that presented the peptide in an individual by using immunogenicity data of the peptide assayed for immunogenicity in affected or vaccinated individuals and selecting the HLA allele that is predicted to have the highest likelihood of presentation by a computational method. For peptides that have not been experimentally assayed, computational presentation predictions can be used.In some embodiments, different computational methods can be combined to predict peptide-HLA immunogenicity or peptide-HLA binding (Liu et al., 2020b). For a given set of peptides and a set of HLA alleles, the term peptide-HLA hit is the number of unique combinations of peptides and HLA alleles that exhibit peptide-HLA immunogenicity or binding at a given threshold. For example, a peptide-HLA hit of 2 may mean that one peptide is predicted to bind to (or induce T cell activation of) two different HLA alleles, or that two peptides are predicted to bind to (or induce T cell activation of) two different HLA alleles, or that two peptides are predicted to bind to (or induce T cell activation of) the same HLA allele. For a given set of peptide and HLA frequencies, HLA haplotype frequencies, or HLA diplotype frequencies, the predicted peptide-HLA hit count is the average number of peptide-HLA hits in each set of HLA representing an individual, weighted by their frequency of occurrence.

[0053] Because immunogenicity can vary between individuals, one way to increase the probability of vaccine efficacy is to use a diverse set of target peptides (e.g., at least two peptides) to increase the likelihood that some subsets of them will be immunogenic in a given individual. Previous studies using mouse models have shown that most MHC-presenting peptides are immunogenic, but immunogenicity varies between individuals, as described by Croft et al. (2019). In some embodiments, experimental peptide-HLA immunogenicity data are used to determine which target peptides and their modifications will be effective immunogens in the vaccine.

[0054] Design considerations for peptide vaccines are outlined in Liu et al., Cell Systems 11, Issue 2, pp. 131-146 (Liu et al., 2020) and (Liu et al., 2020b), as well as in U.S. Patent Nos. 11,058,751 and 11,161,892. These documents are incorporated herein by reference in their entirety.

[0055] Certain target peptides may not bind with high affinity to a wide range of HLA molecules. To increase the binding of target peptides to HLA molecules, their amino acid composition can be modified to alter one or more anchor residues or other residues. In some embodiments, the amino acid composition of a target peptide can be modified to alter one or more residues to increase the immunogenicity of the target peptide when presented by an HLA molecule. The anchor residue is the amino acid that interacts with the HLA molecule and has the greatest influence on the peptide's affinity to the HLA molecule. A peptide with one or more amino acid residues modified is called a heterocritic peptide. In some embodiments, heterocritic peptides include target peptides with residue modifications at the anchor position. In some embodiments, heterocritic peptides include target peptides with residue modifications at non-anchored positions. In some embodiments, heterocritic peptides include target peptides with residue modifications that include non-natural amino acids and / or amino acid derivatives. Modifications for creating heterocritic peptides can improve the binding of the peptide to both MHC class I and MHC class II molecules, and the required modifications may be specific to both the peptide and the MHC class. Peptide anchor residues are less visible to T cell receptors than other peptide residues because they face the MHC molecular groove. Therefore, heterocritic peptides with anchor residue modifications have been observed to induce a T cell response in stimulated T cells that also responds to unmodified peptides. The use of heterocritic peptides in vaccines has been observed to improve vaccine efficacy (Zirlik et al., 2006). In some embodiments, as is well known in the art (Houghton et al., 2007), the immunogenicity of heterocritic peptides is experimentally determined, and their ability to activate T cells that also recognize the corresponding basal (also called seed) peptide is determined.In some embodiments, such assays for the immunogenicity and cross-reactivity of heterocritic peptides are performed when the heterocritic peptide is presented by a specific HLA allele.

[0056] Peptide vaccines that induce immunity against one or more targets In some embodiments, a method is provided for formulating a peptide vaccine using a single vaccine design against one or more targets. In some embodiments, the single target is an exogenous protein with a specific mutation (e.g., KRAS G12D). In some embodiments, the single target is an autologous protein (e.g., a protein overexpressed in tumor cells, such as cancer / testicular antigens). In some embodiments, the single target is a pathogenic protein (e.g., a protein contained in the viral proteome). In some embodiments, multiple targets can be used (e.g., both KRAS G12D and KRAS, or exogenous peptides derived from BCL-ABL transcripts b2a2 and b3a2).

[0057] In some embodiments, the method involves extracting peptides from all target proteome sequences to construct a candidate set, as described in Liu et al. (2020).

[0058] Figures 1 and 2 show flowcharts of exemplary vaccine design methods that can be used for MHC class I or MHC class II vaccine design. The candidate peptide set (see Figures 1 and 2) consists of target peptides extracted by windowing the input protein sequence. In some embodiments, the extracted target peptides are about 8 to about 10 amino acid lengths [e.g., for MHC class I binding (Rist et al., 2013)]. In some embodiments, the extracted target peptides presented by MHC class I molecules are longer than 10 amino acid residues, such as 11 residues (Trolle et al., 2016). In some embodiments, the extracted target peptides are about 13 to about 25 lengths [e.g., for class II binding (Chicz et al., 1992)]. In some embodiments, slide windows of various size ranges described herein are used across the entire proteome. In some embodiments, other target peptide lengths can be used for MHC class I and class II slide windows. In some embodiments, computational predictions of proteasome cleavage are used to filter or select peptides in the candidate set. One computational method for predicting proteasome cleavage is described in Nielsen et al. (2005). In some embodiments, peptide mutation rates, glycosylation, cleavage sites, or other criteria can be used to filter peptides, as described in Liu et al. (2020). In some embodiments, peptides can be filtered based on evolutionary sequence diversity above a predetermined threshold. Evolutionary sequence diversity can be calculated in relation to other species, other pathogens, other pathogen strains, or other related organisms. In some embodiments, the first set of peptides is a candidate set.

[0059] In some embodiments, to design a vaccine against exogenous peptides produced by abnormal gene fusions, target peptides for inclusion in a candidate peptide set are extracted from the gene fusion product, and each extracted target peptide contains a breakpoint between two genes. For example, in some embodiments for designing a BCR-ABL vaccine, the BCR-ABL b3a2(e14a2) and b2a2(e13a2) chimeric protein sequences were obtained from NCBI (GenBank ID CAA10376.1 and CAA10377.1, respectively). For each isoform, slide windows of lengths 8–11 (MHC class I) and 13–25 (MHC class II) were extracted around the BCR-ABL breakpoint. The slide windows can be extracted using the procedures described herein in “MHC Class I Vaccine Design Procedure” and “MHC Class II Vaccine Design Procedure,” in which case P 1...n contains the chimeric protein sequence, t specifies the location of the breakpoint within the chimeric protein sequence, and s=true. In the case of b3a2, the BCR-ABL junction disrupts the triplet codon, generating a novel lysine ("K") at the breakpoint (Clark et al., 2001). In the case of b2a2, codon disruption at the junction causes a change from Asp to Glu, but this novel amino acid is also present at the normal a1a2 junction (Clark et al., 2001). Therefore, in the case of b3a2, all obtained windows spanning the "K" breakpoint were retained. In the case of b2a2, only the window containing the sequence "KEE" was retained, excluding windows found only in the BCR or ABL protein sequences. By applying this procedure to identify breakpoints between fusion genes and using the window strategy described herein, vaccines against other abnormal gene fusions can be generated.

[0060] As shown in Figures 1-2, in some embodiments, the next step of this method involves scoring the target peptides in the candidate set for peptide-HLA binding to all considered HLA alleles, as described in Liu et al. (2020) and Liu et al. (2020b). In some embodiments, the first peptide set is the candidate set after the target peptides have been scored. Scoring can be achieved for human HLA molecules, mouse H-2 molecules, porcine SLA molecules, or MHC molecules of any species for which a predictive algorithm is available or developable. Thus, vaccines targeting non-human species can be designed using this method. Scoring metrics may include affinity of the target peptide to the HLA allele at nanomolar concentrations, eluting ligand, presentation, and other scores or other metrics that can be expressed as percentile rank. The candidate set can be further filtered to exclude peptides whose predictive binding core does not contain the specific pathogenicity or neoantigen target residue of interest, or peptides whose predictive binding core contains the target residue at the anchor position. Furthermore, the candidate set may be filtered for target peptides of specific lengths, such as MHC class I peptides of length 9. In some embodiments, target peptide scoring is achieved by experimental data, or a combination of experimental data and computational prediction methods. If a computational model is not available for making peptide-HLA binding predictions for a particular (peptide, HLA) pair, the binding value for such a pair may be defined by the mean, median, minimum, or maximum immunogenicity values ​​obtained for supported pairs, a fixed value (such as an indicator of non-binding), or by inferring using other techniques that include a function that predicts the most similar (peptide, HLA) pair available in the scoring model.

[0061] In some embodiments, exogenous peptides produced by abnormal gene fusions are not excluded if they contain a fusion breakpoint that falls within the MHC class I or class II anchor position of the HLA allele. For example, in the design of an MHC class I BCL-ABL vaccine, the window is not excluded if the BCR-ABL breakpoint falls within the peptide anchor position. In the case of MHC class II, the scoring model requires that the breakpoint be located within the predicted 9-mer binding core of a given HLA (at any position), and scores for peptide-HLA pairs that do not meet this criterion are excluded. In some embodiments, in the design of an MHC class I BCR-ABL vaccine, the window is excluded if the BCR-ABL breakpoint falls within the peptide anchor position. In some embodiments, in the design of an MHC class II BCR-ABL vaccine, the peptide-HLA score is excluded if the BCR-ABL breakpoint is located within the anchor position of the predicted 9-mer binding core of a given peptide-HLA pair. In some embodiments, for the design of an MHC class II vaccine, the gene fusion breakpoint may be located either inside or outside the predicted 9-mer binding core of a given peptide-HLA pair.

[0062] In some embodiments, a base set (also referred to herein as a seed set) is constructed by selecting peptides from a scoring candidate set using individual peptide-HLA binding or immunogenicity criteria (e.g., a first peptide set) (Figure 1). In some embodiments, since a given peptide has multiple peptide-HLA scores, selection may also be based on the peptide-HLA binding score or peptide-HLA immunogenicity metric that has the best affinity or highest immunogenicity (e.g., predicted to bind most strongly to a given HLA allele or most strongly activate T cells). The criteria used to score peptide-HLA binding during the scoring procedure can correspond to various goals during the base set selection phase and the vaccine design phase. For example, a target peptide with a peptide-HLA binding affinity of 500 nM may be presented by affected individuals, but at a lower frequency than a target peptide with a peptide-HLA binding affinity of 50 nM. During the combinatorial design phase of a vaccine, more constrained affinity criteria, such as 50 nM, can be used to increase the probability that the vaccine peptide will be found and presented by the HLA molecule (e.g., in Figures 1 and 2, a third peptide set, when selecting a vaccine against a target). In some embodiments, a relatively less constrained threshold for peptide-HLA immunogenicity or peptide-HLA binding (e.g., less than approximately 1000 nM or less than approximately 500 nM) is used as a first threshold for filtering candidate peptide-HLA scores for a specific HLA allele score (first peptide scoring and score filtering step in Figures 1 and 2), and a relatively more constrained second threshold (e.g., less than approximately 50 nM) is used for filtering extended set peptide-HLA scores (second peptide filtering and scoring step in Figures 1 and 2). In some embodiments, a specific peptide-HLA score for a given HLA-modified peptide is not used in vaccine design if their unmodified counterpart peptide does not meet a first less restrictive threshold.This filtering of peptide-HLA scores is based on the observation that peptides that are not sufficiently immunogenic for vaccine inclusion may still be antigenic (meeting a first threshold) and therefore potentially recognized by T cell chronotypes expanded by the vaccine. A peptide is antigenic if it is recognized by T cell receptors and elicits a response such as CD8+ T cell cytotoxicity or CD4+ cell activation. Derivatives of antigenic peptides are potently immunogenic and, when included in a vaccine, activate and expand T cells that recognize the antigenic peptide. Expanded T cell proliferation that recognizes unmodified antigenic peptides can provide an immune response that contributes to disease control. In some embodiments, peptides are scored for potential inclusion in a third peptide set (vaccines against targets in Figures 1 and 2) having a peptide-HLA binding affinity of less than approximately 500 nM. In some embodiments, peptides with a peptide-HLA binding affinity of less than approximately 1000 nM for at least one HLA allele are selected for the basic set. Alternatively, predictions of peptide-HLA immunogenicity may be used to identify target peptides for inclusion in the basic set. In some embodiments, peptides can be scored for binding to HLA alleles or for peptide-HLA immunogenicity by experimental observation of the immunogenicity of peptides in the context in which they are presented by HLA alleles, or by experimental observation of the binding of peptides to HLA alleles.

[0063] In some embodiments, peptides can be scored for peptide-HLA binding or peptide-HLA immunogenicity using experimental observations of peptide presentation by specific HLA alleles in tumor cells. In some embodiments, peptides can be scored for peptide-HLA binding or peptide-HLA immunogenicity to that HLA allele using experimental observations of peptide presentation by peptide tumor cells. In some embodiments, peptides can be scored for peptide-HLA binding or peptide-HLA immunogenicity using experimental observations of peptide presentation by peptide tumor cells, and the HLA allele for a particular observed peptide is selected from HLA alleles present in the tumor that satisfy the predicted peptide-HLA binding or immunogenicity threshold. In some embodiments, peptide presentation by tumor cells is experimentally determined using mass spectrometry, as described in Bear et al. (2021) or Wang et al. (2019), and these data are used to score peptide-HLA binding or peptide-HLA immunogenicity. In some embodiments, peptide presentation by tumor cells is experimentally determined using mass spectrometry, and this experimental data is used to certify the inclusion of one or more HLA alleles' seed set peptides for the vaccine. In some embodiments, peptide presentation by tumor cells is experimentally determined using mass spectrometry, and this experimental data is used to exclude the peptide's peptide-HLA binding score or peptide-HLA immunogenicity score if the peptide is not observed to be presented by the HLA alleles using display mass spectrometry. In some embodiments, peptide presentation by tumor cells in an individual is experimentally determined using mass spectrometry, and this experimental data is used to certify the inclusion of one or more HLA alleles' seed set peptides in that individual.In some embodiments, the presentation of peptides by tumor cells in an organism is experimentally determined using mass spectrometry, and these experimental data are used to exclude the peptide-HLA binding score or peptide-HLA immunogenicity score of a peptide if its presentation by HLA alleles is not observed by display mass spectrometry. In some embodiments, computational predictions of the immunogenicity of a peptide in the context of HLA allele presentation, such as the methods of Ogishi et al. (2019) or Bulik-Sullivan et al. (2019), can be used for scoring.

[0064] In some embodiments, the peptide-HLA score or peptide-HLA immunogenicity score of a first peptide in a given basic set (seed set) of HLA alleles is excluded and not considered in vaccine design if the wild-type peptide corresponding to the first peptide (e.g., the unmutated native form of that peptide or the corresponding species peptide within a specified sequence edit distance) has the same HLA allele peptide-HLA score or peptide-HLA immunogenicity score within a specified threshold. The threshold may be based on the difference between the scores of the first peptide and the wild-type peptide, the ratio of the scores of the first peptide and the wild-type peptide, the score of the wild-type peptide, or other metrics. The specified threshold may be either greater than or less than a specified value. In some embodiments, the threshold is defined such that the wild-type peptide is not expected to be presented. In some embodiments, if the peptide-HLA score or peptide-HLA immunogenicity score of a first peptide is excluded during vaccine design, all peptide-HLA scores or peptide-HLA immunogenicity scores of its derivatives for the same HLA allele (e.g., heterocritical peptide derivatives) are also excluded and not considered in vaccine design.

[0065] In some embodiments, the method further includes using population HLA haplotype frequencies to run the OptiVax-Robust algorithm described by Liu et al. (2020) on a scored candidate set to construct a basic set of target peptides (also referred to herein as the seed set) (Figure 2). In some embodiments, HLA diplotype frequencies may be provided to OptiVax. OptiVax-Robust includes an algorithm to eliminate peptide redundancy arising from sliding window techniques with various window sizes, but other redundancy elimination means can be used to enforce a minimum edit distance constraint between target peptides in the candidate set. The seed set size is determined by the diminishing return point of population coverage as a function of the number of target peptides in the seed set. Other criteria may also be used, including the minimum number of vaccine target peptides, the maximum number of vaccine target peptides, and the desired predicted population coverage. In some embodiments, a given population coverage rate is less than approximately 0.4, approximately 0.4–0.5, approximately 0.5–0.6, approximately 0.6–0.7, approximately 0.7–0.8, approximately 0.8–0.9, or greater than approximately 0.9. Another possible criterion is the minimum expected peptide-HLA binding hit count in each individual. In alternative embodiments, the method further includes performing the OptiVax-Unlinked algorithm described by Liu et al. (2020) instead of OptiVax-Robust.

[0066] The OptiVax-Robust method uses binary predictions of peptide-HLA immunogenicity, which can be generated as described in Liu et al. (2020b). The OptiVax-Unlinked method uses the probability that a target peptide binds to an HLA allele, which can be generated as described in Liu et al. (2020). In some embodiments, OptiVax-Unlinked and EvalVax-Unlinked are used in conjunction with peptide-HLA immunogenicity probabilities. Either method can be used for the purposes described herein, and therefore the term "OptiVax" refers to either the Robust method or the Unlinked method. In some embodiments, the peptide-HLA immunogenicity probabilities observed in experimental assays can be used as peptide-HLA binding probabilities in EvalVax-Unlinked and OptiVax-Unlinked. In some embodiments, the population HLA haplotypes or HLA allele frequencies provided to OptiVax for vaccine design represent a global population. In alternative embodiments, the HLA haplotypes or HLA allele frequencies of a population provided to OptiVax for vaccine design are geographically specific. In alternative embodiments, the HLA haplotypes or HLA allele frequencies of a population provided to OptiVax for vaccine design are ancestral specific. In alternative embodiments, the HLA haplotypes or HLA allele frequencies of a population provided to OptiVax for vaccine design are race specific. In alternative embodiments, the HLA haplotypes or HLA allele frequencies of a population provided to OptiVax for vaccine design are specific to individuals with risk factors such as genetic indicators of risk, age, exposure to chemicals, alcohol consumption, chronic inflammation, diet, hormones, immunosuppression, infectious agents, obesity, radiation, sunlight, or smoking. In alternative embodiments, the HLA haplotypes or HLA allele frequencies of a population provided to OptiVax for vaccine design are specific to individuals with a particular HLA allele.In an alternative embodiment, the HLA diplotype provided to OptiVax for vaccine design represents a single individual and is used for the design of personalized vaccines.

[0067] In some embodiments, the base (or seed) set of target peptides (e.g., the first peptide set) resulting from the application of OptiVax to a candidate set of target peptides represents a set of unmodified target peptides that constitute a possible small vaccine design (seed set in Figure 2). The base peptides are the target peptides included in the base or seed peptide set (e.g., the first peptide set). In some embodiments, the seed set (e.g., the first peptide set) is based on filtering candidate peptide scores by predicted or observed affinity or immunogenicity with respect to HLA molecules (seed set in Figure 1). However, to improve the presentation of target peptides across a wide range of HLA haplotypes as much as possible, some embodiments include modification of the seed (or base) set. In some embodiments, experimental assays can be used to ensure that the modified seed (or base) peptides activate T cells that also recognize the base / seed peptides.

[0068] For a given target peptide, optimal anchor residue selection may depend on the HLA allele that binds to and presents the target peptide, and the class of the HLA allele (MHC class I or class II). A seed peptide set (e.g., a first peptide set) can be expanded by including peptides with anchor residue modifications of either MHC class I or II peptides (Figures 1-2). Therefore, one aspect of vaccine design is to consider a method for selecting a limited set of heterocritic peptides derived from the same target peptide for vaccine inclusion, given that different heterocritic peptides will have different and potentially overlapping population coverage rates.

[0069] In some embodiments, all possible anchor modifications for each basic set of target peptides are considered. Typically, peptides to which MHC class I molecules bind have two anchor residues, typically at positions 2 and 9 for 9-mer peptides. In some embodiments, anchors for 8-mer, 10-mer, and 11-mer peptides are found at positions 2 and n, where n is the last position (positions 8, 10, and 11, respectively). For MHC class I molecules, the last position n is referred to herein as the carboxyl terminus "C" position. In some embodiments, 20 possible amino acids are tried at each anchor position to select the best heterocritic peptide. Thus, for MHC class I binding, 400 (i.e., 20 amino acids per two positions = 20) are tried for each basic target peptide. 2 )-1 heterocritical peptide is generated. Typically, the peptide to which the MHC class II molecule binds has four anchor residues, typically located at positions 1, 4, 6, and 9 of the 9-mer binding core. Therefore, in the case of MHC class II binding, 160,000 (i.e., 20 amino acids for each of the four positions = 20) are generated for each basic target peptide. 4 )-1 heterocritic peptide is generated. In some embodiments, more than two (MHC class I) or more than four (MHC class II) positions are considered as anchors. Other methods, including Bayesian optimization, can be used to select the optimal anchor residues and produce heterocritic peptides from each seed (or base) set peptide. Other methods for selecting the optimal anchor residues are described in "Machine learning optimization of peptides for presentation by class II MHCs" by Dai et al. (2020). This literature is incorporated in its entirety herein. In some embodiments, the anchor positions are determined by the HLA allele presenting the peptide, and therefore the set of heterocritic peptides includes all possible anchor modifications for each set of HLA-specific anchor positions.

[0070] In some embodiments, for all target peptides in the base / seed set, new peptide sequences are generated with all possible anchor residue modifications (e.g., MHC class I or class II), resulting in a new heterocritic base set (extended set in Figures 1-2) containing all such modifications. In some embodiments, if one or more of the peptide's anchor residue positions contain a substitutional mutation that distinguishes the peptide from its own peptide, the peptide's anchor residue modification is not included in the heterocritic base set. In some embodiments, only peptide positions that do not contain substitutional mutations that distinguish the base / seed peptide from its own peptide are included in the heterocritic base set. In some embodiments, the peptide's anchor residue modification is not included in the heterocritic base set if one or more peptide mutations have occurred between its adjacent pairs of anchor residues. In some embodiments, for all peptides in the base / seed set, new peptide sequences are generated with anchor residue modifications (e.g., MHC class I or class II), resulting in a new heterocritic base set (extended set in Figures 1-2) containing selected modifications. In some embodiments, the anchor residue positions used to modify the peptide are selected from anchor residue positions determined by the HLA alleles considered during vaccine evaluation. In some embodiments, the heterocritic base set (extended set in Figures 1-2) also includes the original seed (or base) set (seed peptide set in Figures 1-2). In some embodiments, the heterocritic base set includes amino acid substitutions at non-anchored residues. In some embodiments, modifications of the base peptide residues are achieved to alter binding to the T cell receptor and improve therapeutic efficacy (Candia et al., 2016). In some embodiments, the heterocritic base set includes amino acid substitutions of non-natural amino acid analogs.The heterocritic base set is scored for HLA affinity, peptide-HLA immunogenicity, or other metrics described herein (separate rounds of peptide scoring and score filtering are shown in Figures 1 and 2). In some embodiments, the scoring predictions for heterocritic peptide and HLA allele pairs can be further updated to exclude pairs having a seed (or base) peptide from which the heterocritic peptide is derived, for which the heterocritic peptide is not predicted to be presented by the HLA allele at a specified threshold for the peptide-HLA binding score or a specified peptide-HLA immunogenicity metric. Also in some embodiments, the peptide-HLA score can be filtered to ensure that the predicted binding core of the heterocritic peptide presented by a particular HLA allele is precisely aligned with the binding core of the corresponding seed (or base) set target peptide of that HLA allele. In some embodiments, the scoring predictions can be filtered for HLA alleles to ensure that the heterocritic peptide considered with respect to that HLA allele is modified only at the anchor position determined by that HLA allele. The scoring generates peptide-HLA immunogenicity metrics for peptides and HLA alleles, which may be binary, probability of immunogenicity, or other immunogenicity metrics such as peptide-HLA affinity or percentage rank, and may be based on computational prediction, experimental observation, or a combination of both. In some embodiments, the probability of peptide-HLA immunogenicity is used by OptiVax-Unlinked. In some embodiments, heterocritic peptides are included in experimental assays such as MIRA (Klinger et al., 2015) or ELISPOT to determine their peptide-HLA immunogenicity metrics for specific HLA alleles. In some embodiments, MIRA data for heterocritic peptides can be incorporated into the peptide-HLA immunogenicity model using the method of Liu et al. (2020b).In some embodiments, the peptide-HLA immunogenicity metric of the heterocritic peptide is experimentally determined, and its ability to activate T cells that also recognize the corresponding seed (or basal) peptide is performed as is known in the art to certify the vaccine inclusion of the heterocritic peptide (Houghton et al., 2007). In some embodiments, such assays of the immunogenicity and cross-reactivity of the heterocritic peptide are performed when the heterocritic peptide is presented by a specific HLA allele.

[0071] In some embodiments, peptides can be scored for peptide-HLA binding or peptide-HLA immunogenicity using experimental observations of the presentation of heterocritic peptides by specific HLA alleles in cells. In some embodiments, the presentation of heterocritic peptides by cells is experimentally determined using mass spectrometry, as described in Bear et al. (2021) or Wang et al. (2019), and such data is used to score peptide-HLA binding or peptide-HLA immunogenicity. In some embodiments, the presentation of heterocritic peptides by cells is experimentally determined using mass spectrometry, and such experimental data is used to certify the inclusion of heterocritic peptides for inclusion in vaccines. In some embodiments, the presentation of peptides by tumor cells is experimentally determined using mass spectrometry, and such experimental data is used to exclude the peptide's peptide-HLA binding score or peptide-HLA immunogenicity score if the presentation of the peptide by HLA alleles is not observed by display mass spectrometry. In some embodiments, presentation of heterocritic peptides by cells possessing HLA alleles found in an individual is experimentally determined using mass spectrometry, and such experimental data is used to determine the inclusion of heterocritic peptides for inclusion in a vaccine for that individual. In some embodiments, presentation of peptides by tumor cells in an individual is experimentally determined using mass spectrometry, and such experimental data is used to exclude the peptide's peptide-HLA binding score or peptide-HLA immunogenicity score if presentation by HLA alleles is not observed by display mass spectrometry. In some embodiments, computational predictions of the immunogenicity of heterocritic peptides in the context of HLA allele presentation, such as the methods of Ogishi et al. (2019) or Bulik-Sullivan et al. (2019), can be used for scoring.

[0072] In some embodiments, peptides in the heterocritic basic set are removed if (1) one of its anchor sites to the HLA alleles corresponds to a mutation site in the basic / seed peptide from which it originated, which distinguishes the basic / seed peptide from the self-peptide, and (2) the peptide-HLA binding or peptide-HLA immunogenicity of the self-peptide is stronger than a certain threshold for self-peptide binding or immunogenicity. This eliminates peptides in the heterocritic basic set that could cross-react with the self-peptide as a result of sharing residues facing the TCR with the self-peptide. In some embodiments, the threshold for self-peptide binding is approximately 500 nM to 1000 nM.

[0073] In some embodiments, redundant peptides are removed from the heterocritic base set. In some embodiments, the redundant peptide is a first heterocritic peptide that has a lower peptide-HLA immunogenicity score or peptide-HLA binding score for all scored HLAs than a second heterocritic peptide in the heterocritic base set, and both the first and second heterocritic peptides are derived from the same base (or seed) peptide. In some embodiments, peptide redundancy is determined solely by comparing the peptide-HLA immunogenicity score or peptide-HLA binding score of the HLA alleles, and the peptide-HLA immunogenicity score or peptide-HLA binding score of both peptides for the HLA alleles is higher than a predetermined threshold (e.g., 50 nM for binding). In some embodiments, the redundant peptide is a first heterocritic peptide having an average peptide-HLA immunogenicity score or peptide-HLA binding score that is lower in immunogenicity than the average peptide-HLA immunogenicity score or peptide-HLA binding score of the second heterocritic peptide in the heterocritic base set, where both the first and second heterocritic peptides originate from the same base (or seed) peptide, the average scores for the HLA alleles are calculated, and the peptide-HLA immunogenicity scores or peptide-HLA binding scores of both peptides for the HLA alleles are higher in immunogenicity than a predetermined threshold (e.g., 50 nM for binding).In some embodiments, the redundant peptide is a first heterocritic peptide having a weighted peptide-HLA immunogenicity score or peptide-HLA binding score that is lower in immunogenicity than the weighted peptide-HLA immunogenicity score or peptide-HLA binding score of a second heterocritic peptide in the heterocritic base set, where both the first and second heterocritic peptides originate from the same base (or seed) peptide, the weighting is determined by the frequency of HLA alleles in the human population, the weighted score of the HLA allele is calculated, and the peptide-HLA immunogenicity score or peptide-HLA binding score of both peptides against the HLA allele is higher in immunogenicity than a predetermined threshold (e.g., 50 nM for binding).

[0074] In some embodiments, the next step includes scoring a heterocritic base set (a second peptide set) and filtering the resulting scores to create a second peptide set by comparing the peptide-HLA immunogenicity score or peptide-HLA binding score of the peptides against one or more HLA alleles to a threshold. In some embodiments, an affinity criterion of about 50 nM is used to increase the probability that the vaccine peptide will be found and presented by the HLA molecule. In some embodiments, the affinity criterion is more restrictive than 50 nM (i.e., <50 nM). In some embodiments, the affinity criterion is more restrictive than 500 nM (i.e., <500 nM). In some embodiments, individual peptide-HLA binding scores or immunogenicity metrics are determined, and therefore peptides can be retained as long as they meet a criterion for at least one HLA allele, and only peptide-HLA scores that meet the criterion are considered for vaccine design.

[0075] In some embodiments, the next step involves inputting a second peptide set into OptiVax to select a small set of vaccine peptides that maximizes predicted vaccine performance (vaccine performance optimization; Figures 1-2). In some embodiments, predicted vaccine performance is a function of predicted peptide-HLA binding affinity (e.g., a function of the distribution of peptide-HLA binding affinity, either across all peptide-HLA combinations of a given peptide set or weighted by the appearance of HLA alleles in a population or individual). In some embodiments, predicted vaccine performance is the predicted population coverage of the vaccine. In some embodiments, predicted vaccine performance is the predicted number of peptide-HLA hits produced by the vaccine in a population or individual. In some embodiments, predicted vaccine performance requires a minimum predicted number of peptide-HLA hits produced by the vaccine (e.g., 1, 2, 3, 4, 5, 6, 7, 8, or more). In some embodiments, predicted vaccine performance is a function of population coverage and a desired number of predicted peptide-HLA hits produced by the vaccine. In some embodiments, predicted vaccine performance is a metric describing the overall immunogenicity of the vaccine, where all peptides in the vaccine are scored for peptide-HLA immunogenicity against two or more HLA alleles (e.g., three or more HLA alleles). In some embodiments, predicted vaccine performance excludes immunogenic contributions from selected HLA alleles that exceed the maximum peptide-HLA hit count (e.g., 1, 2, 3, 4, 5, 6, 7, 8, or more). In some embodiments, predicted vaccine performance excludes epidemicogenic contributions from individual HLA diplotypes that exceed the maximum peptide-HLA hit count (e.g., 1, 2, 3, 4, 5, 6, 7, 8, or more). In some embodiments, the predicted vaccine performance is the fraction of alleles covered, which is the predicted fraction of HLA alleles in each individual having the minimum number of peptides with predicted peptide-HLA immunogenicity produced by the vaccine (e.g., 1, 2, 3, 4, 5, 6, 7, 8, or more).In some embodiments, predicted vaccine performance is the predicted fraction of HLA alleles in a single individual having the minimum number of peptides (e.g., 1, 2, 3, 4, 5, 6, 7, 8, or more) that have predicted peptide-HLA immunogenicity produced by the vaccine.

[0076] In some embodiments, the vaccine is designed by iteratively selecting peptides from a heterocritical base set (also called an extended set, as shown in Figures 1-2) using criteria that become progressively less stringent regarding predicted peptide immunogenicity or presentation. In some embodiments, peptides are retained if at least one of their peptide-HLA scores is not excluded by the threshold used. In some embodiments, OptiVax is first used to design a vaccine with desired vaccine performance using specific peptide acceptance criteria (e.g., seed HLA-peptide scores from the candidate set must bind to at least one MHC molecule at 500 nM or more strongly, and peptide-HLA scores from the extended set must bind to at least one MHC molecule at 50 nM or more strongly). Next, the vaccine obtained from this application of OptiVax is used as the basis for vaccine augmentation using less stringent criteria (for example, seed peptide-HLA scores from the candidate set must bind to at least one MHC molecule at 1000 nM or more strongly, and peptide-HLA scores from the expansion set must bind to at least one MHC molecule at 100 nM or more strongly) to further improve the desired vaccine performance. The method of vaccine augmentation is described in Liu et al. (2020b), which is incorporated herein by reference in its entirety. In some embodiments, multiple rounds of vaccine augmentation may be used. In some embodiments, the final augmented vaccine is the selected vaccine.

[0077] In some embodiments, the selection of a peptide set that satisfies the desired predicted vaccine performance can be achieved by computational algorithms other than OptiVax. In some embodiments, integer linear programming or mixed integer linear programming is used instead of OptiVax to select the peptide set. An example of integer programming for peptide set selection is described in Toussaint et al. (2008), which is incorporated herein by reference in its entirety. An exemplary solution for mixed integer linear programming is Python-MIP, which can be used in conjunction with Toussaint et al. (2008). A second example of a method for vaccine peptide selection is described in "Maximum n-times Coverage for Vaccine Design" by Liu et al. (2021), which is incorporated herein by reference in its entirety.

[0078] Predictive vaccine performance refers to a metric. Predictive vaccine performance can be expressed as a single number, multiple numbers, any number of non-numbers, or combinations thereof. One or more values ​​can be expressed in any mathematical or symbolic terminology and on any scale (e.g., nominal scale, ordinal scale, interval scale, or ratio scale).

[0079] A seed (or basic) peptide and all modified peptides derived from it constitute a single peptide family. In some embodiments, the maximum number of peptides belonging to the same peptide family (e.g., 1, 2, 3, 4, 5, 6, 7, 8, or more) in the components of peptide-HLA immunogenicity-based vaccine performance for a given HLA allele is given a computational immunogenicity credit for that HLA allele. This limitation on the immunogenicity of peptide families limits the credit caused by multiple modified forms of the same basic peptide. In some embodiments, the methods described herein are included for running OptiVax with an EvalVax objective function corresponding to a desired metric of predicted vaccine performance. In some embodiments, population coverage means the proportion of the target population that presents one or more immunogenic peptides that activate T cells in response to a seed (or basic) target peptide. The metric for population coverage is calculated using HLA haplotype frequencies in a given population, such as a representative human population. In some embodiments, the population coverage metric is calculated using marginal HLA frequencies in the population. Maximizing population coverage means selecting a set of peptides (either a basic peptide set, a modified peptide set, or a combination of basic and modified peptides; e.g., a first peptide set, a second peptide set, or a third peptide set) that collectively yields the maximum fraction of the population exhibiting at least a minimum number (e.g., 1, 2, 3, 4, 5, 6, 7, 8, or more) of immunogenic peptide-HLA bindings, based on the proportion of HLA haplotypes in a given population (e.g., a representative human population). In some embodiments, this process includes OptiVax selection of heterocritical peptides (such as those described herein) that activate T cells in response to the corresponding seed (or basic) peptide and heterocritical basic peptide to improve population coverage. In some embodiments, the final vaccine design always includes a seed (or basic) target peptide.In some embodiments, a peptide is considered a candidate for vaccine design only if it has been observed to be immunogenic in clinical data, animal models, or tissue culture models (e.g., included in the first, second, and / or third peptide set).

[0080] While heterocritical peptides are used as exemplary embodiments in this disclosure, any modified peptide can be used instead of heterocritical peptides. A modified peptide is a peptide having one or more amino acid substitutions of the target base / seed peptide. The amino acid substitutions may be located at anchor positions or at any other non-anchored positions.

[0081] In some embodiments, a candidate vaccine peptide (e.g., a basic peptide or a modified peptide) is excluded from vaccine inclusion if it activates T cells that recognize the self-peptide (e.g., this can be achieved in the first and / or second rounds of peptide filtering and selection shown in Figures 1 and 2). In some embodiments, a candidate vaccine peptide (e.g., a basic peptide or a modified peptide) is computationally excluded from vaccine inclusion if, upon binding to an HLA allele, its outward-facing amino acids are similar to outward-facing self-peptide residues presented by the same HLA allele. In this case, similarity may be defined by identity or by a similarity metric such as the BLOSUM matrix (the BLOSUM matrix is ​​known in the art). Testing of vaccine peptides for their ability to activate T cells that recognize the self-peptide can be achieved experimentally by vaccinating animal models and subsequently performing ELISPOT or other immunogenicity assays, or by using human tissue protocols. In any case, models having HLA alleles that present the vaccine peptide are used. In some embodiments, human primary blood mononuclear cells (PBMCs) are stimulated with a vaccine peptide to enable T cell proliferation, and then T cell activation by the self-peptide is assayed as described in Tapia-Calle et al. (2019) or by other methods known in the art. In some embodiments, if T cells are activated by the self-peptide, the vaccine peptide is excluded from vaccine inclusion. In some embodiments, computational predictions of the ability of peptides to activate T cells that also recognize the self-peptide can be used. Such predictions may be based on modeling of the outward-facing residues of the peptide-HLA complex and their interactions with other peptide residues.In some embodiments, a candidate vaccine peptide (e.g., a basic peptide or a modified peptide) is excluded from vaccine inclusion or experimentally tested for cross-reactivity if it is predicted to activate T cells that also recognize the self-peptide based on the structural similarity between the peptide-MHC complex of the candidate peptide (e.g., basic peptide or modified peptide) and the peptide-MHC complex of the self-peptide. One method for predicting peptide-MHC structure is described in Park et al. (2013).

[0082] In some embodiments, candidate heterocritic vaccine peptides (e.g., modified peptides) and the peptide-HLA binding score or peptide-HLA immunogenicity metric of the HLA allele are excluded from consideration during vaccine design if the candidate heterocritic vaccine peptide does not activate T cells that recognize its corresponding basal / seed target peptide for a given HLA allele (second round of peptide scoring and score filtering, Figures 1-2). Testing of candidate heterocritic peptides (e.g., modified peptides) for their ability to activate T cells that recognize their corresponding seed (or basal) target peptide for the same HLA allele can be experimentally achieved by vaccinating animal models followed by ELISPOT or other immunogenicity assays, or by using human tissue protocols. In any case, a model having HLA alleles that present heterocritic peptides is used. In some embodiments, human PBMCs are stimulated with heterocritic peptides to enable T cell proliferation, and then T cell activation by seed (or basic) target peptides is assayed as described in Tapia-Calle et al. (2019) or using other methods known in the art. In some embodiments, computational predictions of the ability of heterocritic peptides to activate T cells that also recognize the corresponding seed (or basic) target peptides can be used. Such predictions may be based on modeling of outward-facing residues of the peptide-HLA complex and their interactions with other peptide residues.In some embodiments, the structural similarity between the peptide-HLA complex of the heterocritic peptide and the peptide-HLA complex of the corresponding seed (or basic) target is used to certify the heterocritic peptide for vaccine inclusion, or experimental immunogenicity testing may be required before vaccine inclusion.

[0083] TCR Interface Divergence (TCRID) is the least-squares mean square deviation of the difference between the 3D position of the TCR opposing residue of a first peptide and the corresponding residue position of a second peptide, with respect to a particular HLA allele. In some embodiments, other metrics are used for TCRID instead of the least-squares mean square deviation. In some embodiments, other metrics are used for TCRID, including positional deviations of non-TCR opposing residues and MHC residues derived from a particular HLA allele. In some embodiments, TCRID is used to predict whether two peptides will activate the same T cell chronotype when presented by a given HLA allele. In some embodiments, the TCRID metric for a pair of peptides can be calculated considering a given HLA molecule using FlexPepDock (London et al., 2011, this document is incorporated herein by reference in its entirety) or DINC (Antunes et al., 2018, this document is incorporated herein by reference in its entirety), in conjunction with the crystal structure of the HLA molecule. In some embodiments, the TCRID is calculated by (1) determining the 3D peptide-HLA structures of two different peptides to which a specific HLA allele binds, (2) aligning the HLA alpha helix of the peptide-HLA structures, and (3) calculating the least squares mean square deviation of the difference between the TCR opposing residues of the two peptides with respect to the aligned alpha helix reference frame.

[0084] In some embodiments, the second peptide scoring and score filtering step in Figures 1 and 2 eliminates the peptide-HLA binding or immunogenicity score of a heterocritic peptide for a particular HLA allele if the HLA-specific TCRID between the heterocritic peptide and its corresponding base (or seed) peptide from which it is derived exceeds a first TCRID threshold. In some embodiments, the second peptide scoring and score filtering step in Figures 1 and 2 eliminates all peptide-HLA binding or immunogenicity scores of a heterocritic peptide if the HLA-specific TCRID between the heterocritic peptide and its corresponding unmutated self-peptide from which it is derived falls below a second TCRID threshold. In some embodiments, the first peptide scoring and score filtering step in Figures 1 and 2 eliminates all peptide-HLA binding or immunogenicity scores of a candidate peptide if the HLA-specific TCRID between the peptide and its corresponding unmutated self-peptide falls below a third TCRID threshold. In some embodiments, any of the TCRID thresholds are determined by experimentally observing or computationally predicting the cross-reactivity of TCR molecules to peptide-HLA complexes.

[0085] Figures 3 (MHC Class I) and 4 (MHC Class II) show the predicted population coverage of OptiVax-Robust-selected single-target specific vaccines with different numbers of peptides, designed for KRAS mutations G12D, G12V, G12R, G12C, and G13D. Figures 3 and 4 show that the predicted population coverage increases as the number of peptides in the vaccine increases. The population coverage shown in Figures 3 and 4 is for individuals with the specific mutations that the vaccine is designed to cover. As the peptide count increases, typically the average number of peptide-HLA hits in each individual of that population also increases.

[0086] Figure 5 shows the predicted population coverage by vaccine size for MHC class I vaccines against BCL-ABL fusions that produce b3a2. Figure 6 shows the predicted population coverage by vaccine size for MHC class II vaccines against BCL-ABL fusions that produce b3a2. Figure 7 shows the predicted population coverage by vaccine size for MHC class I vaccines against BCL-ABL fusions that produce b2a2. Figure 8 shows the predicted population coverage by vaccine size for MHC class II vaccines against BCL-ABL fusions that produce b2a2.

[0087] Using OptiVax, vaccines can be designed to maximize the fraction / proportion of the population in which HLA molecules are predicted to bind to and present at least p peptides in the vaccine. In some embodiments, this prediction (e.g., scoring) includes experimental immunogenicity data to directly predict that at least p peptides are immunogenic. A numerical value p can be input into OptiVax, and OptiVax can be run multiple times with varying values ​​of p to obtain a predicted optimal target peptide set for various peptide counts p. Larger values ​​of p will increase vaccine redundancy at the cost of more peptides to achieve the desired population coverage. In some embodiments, given a particular heterocritical base set, it may not be possible to achieve a given population coverage. In some embodiments, the numerical value p is a function of the desired size of the vaccine.

[0088] Using the methods described herein, separate vaccine formulations can be designed for immunity based on MHC class I and class II.

[0089] In some embodiments, this procedure is used to create a vaccine for an individual. In some embodiments, the target peptide present in the individual is determined by sequencing the individual's tumor RNA or DNA and identifying mutations that produce an exogenous peptide. One embodiment of this method is described in U.S. Patent No. 10,738,355, which is incorporated herein by reference in its entirety. In some embodiments, a peptide sequencing method is used to identify the target peptide in the individual. One embodiment of this is described in U.S. Patent Application Publication No. 2011 / 0257890. In some embodiments, the target peptide used for a vaccine in an individual is selected if an RNA encoding a self-peptide, exogenous peptide, pathogenic peptide, or self-peptide, exogenous peptide, or pathogenic peptide is observed and present at a predetermined level in a sample derived from the individual. The target peptide in the individual is used to construct the vaccine disclosed herein. For vaccine design, OptiVax provides a diplotype containing the individual's HLA type. In an alternative embodiment, an individual's HLA type is separated into multiple diplotypes having a combined frequency of 1, each diplotype containing one or more HLA alleles of the individual, and a notation indicating that other allele locations should not be evaluated. The use of multiple diplotypes leads to an increased likelihood that the OptiVax objective function will result in immunogenic peptides being presented by all constructed diplotypes. This achieves the objective of maximizing the number of distinct HLA alleles in individuals exhibiting peptide-HLA immunogenicity, and thus improving the allele coverage of the vaccine in the individual.

[0090] Figure 9 shows the predicted vaccine performance (predicted peptide-HLA hit count) of 10 G12V MHC class I vaccines for a single individual with MHC class I HLA diplotypes HLA-A02:03, HLA-A11:01, HLA-B55:02, HLA-B58:01, HLA-C03:02, and HLA-C03:03. Using OptiVax, 10 G12V MHC class I vaccines against these HLA diplotypes were designed with peptide counts ranging from 1 to 10. In the case of the results in Figure 9, OptiVax was run using six synthetic diplotypes, each equally weighted and each having one HLA allele derived from the individual HLA diplotype, with other allele positions not evaluated. The ten peptide vaccines in Figure 9 include SEQ ID NOs: 3 (GAVGVGKSL), 4 (LMVVGAVGV), 7 (VVGAVGVGK), 14 (GPVGVGKSV), 69 (LMVVGAVGI), 72 (LMVVGAVGL), 131 (GAVGVGKSM), 138 (GPVGVGKSA), 142 (VTGAVGVGK), and 198 (VAGAVGVGM). Two peptides, SEQ ID NOs: 3 (GAVGVGKSL) and 131 (GAVGVGKSM), are predicted to bind to two HLA alleles with an affinity of 50 nM or less.

[0091] MHC Class I Vaccine Design Procedure In some embodiments, the MHC class I vaccine design procedure consists of the following calculation steps.

[0092] In some embodiments, the inputs for the calculation are as follows: P 1...n : A peptide sequence (length n) containing the target neoantigen or pathogenic target (e.g., KRAS G12D, KRAS G12V, KRAS G12R, KRAS G12C, KRAS G13D, BCR-ABL b3a2, BCR-ABL b2a2). i This indicates the amino acid at position i. t: The location of the target mutation in P, t∈[1,...n] (for example, in the case of KRAS G12D, t=12). s: Substitutional mutation s∈[true, false] is true if the mutation is a substitution, and false if the mutation is a deletion or insertion, or if the peptide does not contain a mutation (e.g., a pathogen target). If the mutation is a deletion or insertion, t indicates the position immediately preceding the deletion or insertion. τ1: Threshold for potential peptide presentation by MHC (e.g., 500 nM binding affinity) for peptide-MHC scoring. τ2: Threshold for MHC-mediated prediction of peptides for peptide-MHC scoring (e.g., 50 nM binding affinity) TIFF0007829242000001.tif16159N: Parameters for the EvalVax and OptiVax objective functions. Specifies the minimum predicted individual hit count for the population coverage target for an individual to be considered covered. Default setting = 1 (calculates P(n≧1) population coverage).

[0093] In some embodiments, the peptide-HLA scoring function used is as follows: TIFF0007829242000002.tif52159

[0094] TIFF0007829242000003.tif32157 TIFF0007829242000004.tif33157

[0095] The second condition, j≠{t-(k-1),t-1}, excludes peptides where the mutation at t is at the P2 or Pk position (i.e., the anchor position) of the windowed k-mer peptide, and the mutation is a substitution.

[0096] Therefore, B contains a native peptide that is predicted to be potentially presented by at least one HLA. Create a set B' of all heterocritical peptides derived from the peptide in B: In the formula TIFF0007829242000006.tif16159, ANCHOR-MODIFIED(b) returns the set of all 399 anchor-modified peptides derived from b (having all possible modifications to the amino acids P2 and P9).

[0097] TIFF0007829242000007.tif30159

[0098] In the formula TIFF0007829242000008.tif26159, each heterocritic peptide b'∈B' is a mutation of the base peptide b∈B. This condition compels that if h was not predicted to potentially present b, then all heterocritic peptides b' derived from b will not be presented by h (even if h was otherwise predicted to present b').

[0099] TIFF0007829242000009.tif53158

[0100] In some embodiments, this procedure can be repeated independently for each target of interest, and the resulting independent vaccine sets can be combined into a combination vaccine as described below.

[0101] MHC Class II Vaccine Design Procedure In some embodiments, the MHC class II vaccine design procedure consists of the following calculation steps.

[0102] In some embodiments, the inputs for the calculation are as follows: P 1...n : A peptide sequence (length n) containing the target neoantigen (e.g., KRAS G12D, KRAS G12V, KRAS G12R, KRAS G12C, KRAS G13D, BCR-ABL b3a2, BCR-ABL b2a2). i This indicates the amino acid at position i. t: The location of the target mutation in P, t∈[1,...,n] (for example, in the case of KRAS G12D, t=12). s: Substitutional mutation s∈[true, false] is true if the mutation is a substitution, and false if the mutation is a deletion or insertion, or if the peptide does not contain a mutation (e.g., a pathogen target). If the mutation is a deletion or insertion, t indicates the position immediately preceding the deletion or insertion. τ1: Threshold for potential peptide presentation by MHC (e.g., 500 nM binding affinity) for peptide-MHC scoring. τ2: Threshold for MHC-mediated prediction of peptides for peptide-MHC scoring (e.g., 50 nM binding affinity) TIFF0007829242000010.tif21159N: Parameters for the EvalVax and OptiVax objective functions. Specifies the minimum predicted individual hit count for the population coverage target for an individual to be considered covered. Default setting = 1 (calculates P(n≧1) population coverage).

[0103] In some embodiments, the peptide-HLA scoring function used is as follows: TIFF0007829242000011.tif26158 TIFF0007829242000012.tif27158 TIFF0007829242000013.tif15158

[0104] TIFF0007829242000014.tif28158 TIFF0007829242000015.tif29158 TIFF0007829242000016.tif21158

[0105] TIFF0007829242000017.tif41158 Note that S1 is a binary matrix, where 1 indicates that the HLA is predicted to potentially present the peptide, and 0 indicates that there is no potential presentation.

[0106] TIFF0007829242000018.tif31158

[0107] In the equation of TIFF0007829242000019.tif21159, P t is the target residue of interest (e.g., the mutation site of KRAS G12D). This condition enforces that for all (peptide, HLA allele) pairs having a non-zero score in S2, the target residue falls within the binding core at non-anchor positions, allowing the binding core to vary by allele for each peptide (the binding core for a particular peptide may vary based on the HLA allele presenting that peptide). Thus, for each pair (p, h), if the predicted binding core C[p, h] specifies the target residue P t at an anchor position (P1, P4, P6, or P9 of the 9-mer core), or if P t is not contained within the binding core core, then S2[p, h]=0. In an alternative embodiment, P t may be located outside the core or at a non-anchor position within the core. In some embodiments, P t may be located only at specific positions inside and / or outside the core. In some embodiments, the binding core prediction in C is accompanied by a prediction confidence. In some embodiments, if the confidence of the predicted core C[p, h] is below a desired threshold (e.g., 0.5, 0.6, 0.7, 0.8, or 0.9), then S2[p, h]=0.

[0108] TIFF0007829242000020.tif21159

[0109] Next, create a set B’ of all heteroclitic peptides derived from the peptides in B: In the equation of TIFF0007829242000021.tif16159, ANCHOR-MODIFIED(b, c) represents all 20 derived from b that have all possible modifications to the amino acids at P1, P4, P6, and P9 of the 9-mer binding core c 4- Returns a set of 1 anchor-modified peptides. Thus, for each basic peptide b, heterocritic set B' contains all anchor-modified peptides b' that have modifications to all intrinsic cores of b identified for any HLA allele that potentially presents b with valid core positions indicated by the scoring matrix S2.

[0110] TIFF0007829242000022.tif33159

[0111] TIFF0007829242000023.tif28159

[0112] Conditional on identified binding cores of heterocritical and basal peptides arising at the same offset by a specific HLA, update scoring matrix for heterocritical peptides. Calculate TIFF0007829242000024.tif16157. In the formula, each heterocritical peptide b'∈B' is a mutation of the base peptide b∈B. This condition forces the binding core of the heterocritical peptide b' to be in the same relative position as the base peptide b, implicitly targeting the target residue P. t This forces the 9-mer bonded core to remain in a non-anchored position (step 3).

[0113] The updated score matrix for heterocritical peptides is based on the potential presentation of corresponding basic peptides by each HLA: Calculate TIFF0007829242000025.tif16159. In the formula, each heterocritic peptide b'∈B' is a mutation of the base peptide b∈B. This condition compels that if h is not predicted to present b, then all heterocritic peptides b' derived from b will not be presented by h (even if h is otherwise predicted to present b').

[0114] TIFF0007829242000026.tif53159

[0115] In some embodiments, this procedure can be repeated independently for each single target of interest, and the resulting independent vaccine sets can be combined into a combination vaccine as described below.

[0116] A method for designing MHC class I or class II vaccines that prioritizes peptide preservation. In some embodiments, vaccine inclusion prioritizes peptide sequences that are more conserved across the target strain, species, or other protein sources. In some embodiments, vaccine design considers a set of related protein sequences called protein variants. Protein variants are an example of a protein sequence family, and protein variants may be sequences derived from various species, pathogen strains, virus strains, or other variants considered in vaccine design. In some embodiments, each protein variant has a related probability called a protein variant probability, and the sum of all protein variant probabilities for a supply set of protein variants is 1. In some embodiments, multiple target proteins can be considered in the design of a single vaccine using MHC class I or class II vaccine design methods that prioritize peptide conservation. In these embodiments, protein variants of all target proteins are considered collectively to generate candidate peptides. In some embodiments, the sum of protein variant probabilities across all of the multiple proteins considered is 1.

[0117] A set of candidate peptides is generated from each protein variant using a slide window method to parse the protein variant into peptide sequences. In some embodiments, for MHC class I, 8-mer, 9-mer, 10-mer, and 11-mer peptides are generated, but this process can be carried out with any desired window length, and the resulting peptide sets can be combined. In some embodiments, for MHC class I, only 9-mer peptides are generated. In some embodiments, for MHC class II, all windowed peptides of lengths 13–25 are generated, but this process can be carried out using any desired window length (e.g., only 15-mer peptides). In some embodiments, as described in Liu et al. 2020, peptides predicted to be glycosylated in a given protein variant are removed, and that variant is not considered. This literature is incorporated herein by reference in its entirety.

[0118] In some embodiments, the conservation degree of each generated peptide sequence (MHC class I or class II) is defined as the fraction of input protein variants from which that peptide sequence arises. For example, if a given 9-mer peptide sequence is present in peptides generated from 90% of the protein variants provided as input, its conservation degree is 0.90. In some embodiments, the conservation degree of each generated peptide sequence (MHC class I or class II) is defined as the sum of the frequencies of protein variants from which that peptide sequence arises. For example, if a given 9-mer peptide sequence arises in peptides generated from protein variants with protein variant probabilities of 0.10 and 0.20, its conservation degree is 0.30. In some embodiments, this functionality is implemented by a ComputeConservation function that calculates the sum of the frequencies of protein variants containing the peptide sequence. In some embodiments, if there are not enough protein variants to calculate the expected future conservation degree, ComputeConservation can be implemented using a method for predicting conservation, such as those found in Hie et al., 2021, which are incorporated herein by reference in their entirety.

[0119] In some embodiments, conservation is considered in vaccine design by prioritizing the inclusion of peptides with higher conservation than other peptides in order to satisfy the desired vaccine performance metric. In some embodiments, this vaccine design method first attempts to design a vaccine using candidate peptides that all satisfy a first conservation threshold, and if the desired vaccine performance is not met, attempts to satisfy the desired vaccine performance metric by iteratively adding additional peptides with less stringent conservation. In some embodiments, the conservation-prioritizing vaccine design proceeds by setting vaccine design D to an empty set and then performing the following steps: (1) selecting candidate peptides that do not exist in D and each peptide satisfies a conservation threshold for creating a candidate peptide set; (2) selecting vaccine designs with various peptide numbers / combinations from this candidate set and augmenting the vaccine designs included in D by optimizing the vaccine performance metric using the method disclosed herein for MHC class I or class II vaccine designs (one implementation form of vaccine augmentation is (Liu et al., (2021) The literature is incorporated herein by reference in its entirety, (3) From step 2, select a minimum vaccine peptide set design that either satisfies the desired vaccine performance metric or, if one more peptide is added to the selected set, does not result in the desired minimum improvement to the vaccine performance metric; (4) If a vaccine peptide set is found in step 3, add the vaccine peptide set design from step 3 to vaccine design D; and (5) Determine whether vaccine design D satisfies the desired vaccine performance metric target, and if so, return vaccine design D as the final vaccine design.If vaccine design D does not meet the desired vaccine performance metric target in step 6, the calculation proceeds to the following steps: (6) setting the updated conservation threshold lower (less constrained) than the current conservation threshold, and (7) repeating the process starting from step 1, while retaining the current vaccine design D and the current candidate set, until either the desired vaccine performance metric target is achieved in step 6, or the updated conservation threshold is lower than the minimum desired conservation threshold. In either iteration, if the updated conservation threshold is lower than the minimum desired conservation threshold, the latest version of vaccine design D will be used as the final vaccine design. When the process is complete, the final vaccine design D will contain all the peptides that can be used in the vaccine.

[0120] In some embodiments, the MHC class I or class II vaccine design procedure consists of the following calculation steps.

[0121] In some embodiments, the inputs for the calculation are as follows: TIFF0007829242000027.tif19159O j : Protein mutant P j Probability of protein variants t j : Targeted mutant protein variant P j Position t∈[1,...n] D: Vaccine design initialized to the empty set φ s: Substitution mutation s∈[true, false] is true if the mutation is a substitution, and false if the mutation is a deletion or insertion, or if the peptide does not contain a mutation. If the mutation is a deletion or insertion, t indicates the position immediately before the deletion or insertion. c1: Initial conservation level of peptide c c :Current preservation threshold c2: Changes in conservation level in each iteration c m :Minimum final conservation degree ν: Targeted vaccine performance metrics ν d Minimum change in vaccine performance metrics to increase vaccine size N: Parameters for the EvalVax and OptiVax objective functions. Identify the minimum predicted individual hit count for the population coverage target where an individual is considered covered. Initial setting = 1 (calculate P(n≧1) population coverage). TIFF0007829242000028.tif12159

[0122] Protein mutant sequence P j Using this method, windowed peptides are generated across protein sequences with peptide lengths of k residues, starting from each position m. The result is set X, which contains all of the peptide sequences of the protein mutant Pj. j P j,m...m+(k-1) The subscript indicates the specified protein P j Sequences are generated only if they fall within the specified range. In some embodiments, for MHC class I, k is selected to generate 8-mer, 9-mer, 10-mer, and 11-mer, but this process can be carried out with any desired window length and the resulting peptide sets can be combined. In some embodiments, for MHC class I, only 9-mers are generated. In some embodiments, for MHC class II, the inventors extract all window-treated peptides of lengths 13-25, but this process can be carried out using any desired window length (e.g., only 15-mers). TIFF0007829242000029.tif31157

[0123] In some embodiments, for MHC class I, the second condition m≠{t-(k-1),t-1} excludes the peptide if the mutation at t is at the P2 or Pk position (i.e., the anchor position) of the windowed k-mer peptide, the mutation is a substitution, and it is for MHC class I design. For MHC class II design methods, MHC class II anchor positions are filtered out.

[0124] Create set B of all peptides that arise from any input protein variant. TIFF0007829242000030.tif27157

[0125] TIFF0007829242000031.tif20159

[0126] Next, the current preservation threshold is set as the initial preservation threshold. TIFF0007829242000032.tif8157

[0127] TIFF0007829242000033.tif30159

[0128] TIFF0007829242000034.tif99158

[0129] TIFF0007829242000035.tif48158

[0130] TIFF0007829242000036.tif16158

[0131] In step 5, it is determined whether vaccine design D meets the desired vaccine performance metric target. If vaccine design set D meets the final vaccine performance design metric ν, D is returned as the final design.

[0132] In step 6, the conservation threshold is updated to be lower (fewer constraints) than the current conservation threshold. If vaccine design set D does not satisfy the final vaccine performance design metric ν, c c To reduce. TIFF0007829242000037.tif7157

[0133] In step 7, the process starting from step 1 is repeated, while retaining the current vaccine design D and the current candidate set, until either the desired vaccine performance metric target is achieved in step 5, or the updated conservation threshold falls below the minimum desired conservation threshold.c <c m If so, return design set D as the final vaccine. Otherwise, return to process 1 and repeat all subsequent processes.

[0134] In some embodiments, this procedure can be repeated independently for each target pathogen gene variant or target variant, and the resulting independent vaccine sets can be integrated into a combination vaccine.

[0135] Methods of combining multiple vaccines The method described above generates an optimized target peptide set (e.g., a third peptide set) for one or more individual targets. In some embodiments, a method is provided for designing separate vaccines for MHC class I and class II-based immunity against multiple targets (e.g., two or more targets such as KRAS G12D and KRAS G12V).

[0136] In some embodiments, a method for producing combination peptide vaccines against multiple targets is disclosed by using tables of disease presentations based on empirical data from sources such as cancer genome atlases (TCGA). Figure 10 shows one embodiment for factoring disease presentation type probabilities (e.g., pancreatic cancer, colorectal cancer, and skin cancer) by the probabilities of targets presented for each disease presentation against various mutant targets (e.g., KRAS G12D, KRAS G12V, and KRAS G12R). A presentation is a unique set of targets presented by one form of the disease (e.g., different types of cancer or cancer signs as shown in Figure 10). Figure 10 shows, for each presentation, an example of the probability of that presentation and the probability of a given target being observed. For a given presentation, there may be one or more targets, each with a probability. In some embodiments, the method for designing a multi-target vaccine would involve allocating peptide resources to induce disease immunity based on the presentations and their respective target probabilities, as shown in Figure 10, for example. In some embodiments, presentations correspond to the prevalence of targets in different human populations or different risk groups. The probability of a target in a population is calculated by summing the products of the probability of the presentation and the probability of the target in that presentation for each possible presentation. Figure 10 shows the weights used to integrate individual vaccines (rows) for each target into combination vaccines (columns) for each disease symptom. The values ​​represent the observed fraction of cases containing each target mutation. The data are from the Cancer Genome Atlas (TCGA). The TCGA data has been filtered for each disease symptom to include cases where the major site is the symptom.

[0137] In some embodiments, if mutations to different proteins produce identical underlying peptides, then the same vaccine design will be generated for mutations to different proteins. For example, in some embodiments of underlying peptide selection, the following mutations share the same set of underlying peptides and therefore have the same vaccine design: HRAS Q61K, NRAS Q61K, and KRAS Q61K; HRAS Q61L, NRAS Q61L, and KRAS Q61L; HRAS Q61R, NRAS Q61R, and KRAS Q61R. Referring to Figure 10, in some embodiments, if two mutations have identical individual vaccine designs, their presentation specific probabilities are added together when weighting the individual vaccine designs for inclusion in a combination vaccine, as described below (e.g., for thyroid cancer NRAS Q61R and HRAS Q61R).

[0138] Referring to Figure 11, in some embodiments, this method involves first designing individual peptide vaccines for each target to create combination vaccine designs for multiple targets. This initially results in a set of target-specific vaccine designs. In some embodiments, the marginal predicted vaccine performance of each target-specific vaccine of size k is defined by subtracting the predicted vaccine performance of a vaccine of size k-1 from the predicted vaccine performance of a vaccine of size k-1. Since the vaccine composition may change as the number of peptides used in the vaccine increases, the marginal predicted vaccine performance of each target-specific vaccine is used instead of a specific set of peptides to calculate the contribution to the combination vaccine.

[0139] In some embodiments, for each target-specific vaccine size, the weighted limit predicted vaccine performance of the target-specific vaccine design is calculated as shown in Figure 11. For a given target-specific vaccine size, the weighted predicted vaccine performance is calculated by multiplying its predicted vaccine performance by the probability of the target in the population (for example, by using values ​​as shown in Figure 10). The limit weighted predicted vaccine performance of a target-specific vaccine is the weighted coverage at size k minus the coverage at size k-1. The limit weighted predicted vaccine performance of a target-specific vaccine of size 1 is its weighted predicted vaccine performance. As shown in Figure 11, the limit weighted predicted vaccine performances of all vaccines are combined into a single list, and this combination list is sorted from maximum to minimum by the weighted limit predicted vaccine performance of the target-specific vaccines. Then, the combination vaccine of size n is determined by the first n elements of this list. The peptides of the combination vaccine are determined by individual peptide-targeted vaccines whose sizes add up to n, and whose sum of weighted predicted vaccine performances is the same as the sum of the first n elements of the sorted list. This maximizes the predicted vaccine performance of a combination vaccine of size n.

[0140] In some embodiments, a combined multi-target vaccine can be designed based on its overall predicted coverage for diseases described in a presentation table used (see, for example, Figure 10), by its predicted coverage for specific symptoms, and / or by its predicted coverage for specific targets by adjusting the weights used accordingly for predicted vaccine performance. Once the desired coverage level is selected, the peptides of the combined vaccine are determined by the contribution of the target-specific design. For example, if the combined vaccine includes a target-specific vaccine of size k, then a vaccine peptide of size k for this target is used in the combined vaccine.

[0141] As an example of one embodiment, Figure 10 shows the respective probabilities of mutations (e.g., KRAS G12D, G12V, and G12R) and various cancer signs (e.g., pancreatic cancer) occurring in individuals. Figures 3 (MHC class I) and 4 (MHC class II) show the population coverage of target-specific vaccines against the KRAS G12D, G12V, G12R, G12C, and G13D targets using the method for vaccines described herein. The marginal population coverage of each target-specific vaccine at a given vaccine size is its size and the improvement in coverage at size-1. Coverage is zero when the peptide is absent. The marginal coverage of each target-specific vaccine is multiplied by the probability of the target in the population, determined by the proportions shown in Figure 10 for the selected sign (e.g., pancreatic cancer). These weighted marginal coverages for all target-specific vaccines are sorted to determine the best target-specific composition. The resulting list describes the composition of each size k combination vaccine for the selected symptom by taking the first k element of the list. As an example of one embodiment, Figures 12 (MHC class I) and 13 (MHC class II) show the target-specific contributions of each vaccine size of combination KRAS vaccines for three mutant KRAS G12D, G12V, and G12R. In Figures 12 and 13, these examples were calculated using the method of the combination vaccine protocol described herein. For each combination vaccine size, target-specific vaccines of different components are used for the indicated symptom. Table 1 (below) contains peptides present in independent (single-target) and combination (multiple-target) MHC class I vaccine designs for the KRAS G12D, G12V, G12R, G12C, and G13D targets. Table 2 (below) contains peptides present in independent (single-target) MHC class II vaccine designs for KRAS G12D, G12V, G12R, G12C, and G13D targets. Any subset of these individual / single-target vaccines can be combined to create MHC class II vaccines targeting two or more multiple targets.In an alternative embodiment, the sequence listing provides heterocritical peptides useful for MHC class I vaccines targeting KRAS G12D, G12V, G12R, G12C, and G13D.

[0142] Combination vaccine design procedure In some embodiments, the procedures described herein are used to combine individual miniature vaccines optimized for different targets into a single optimized combination vaccine.

[0143] In some embodiments, the calculation input is as follows: τ: A set of target neoantigens or pathogenic targets (e.g., KRAS G12D, KRAS G12V, KRAS G12R, BCR-ABL b3a2, BCR-ABL b2a2) ν: A vaccine set individually optimized for each target. V t,k This represents the optimal vaccine set of exactly k peptides for the target t∈T (calculated, for example, by the procedure described above). t,k+1 This does not necessarily mean ν t,k Please note that this is not necessarily a higher-tier set. W:τ→[0,1]: A target weighting function that maps each target t∈τ to the probability or weight of t in the manifestation of a particular objective (e.g., pancreatic cancer, submission A, see e.g., Table 1). TIFF0007829242000038.tif26159

[0144] In step 1, an optimized vaccine of size 1 to m is set for each target t individually. t,k The calculation is performed as follows: (where k represents the vaccine size), and then the vaccine performance for each vaccine size is calculated. For each target t (individually) and vaccine size (peptide count) k, the unweighted population coverage rate c is calculated. t,k Calculate. TIFF0007829242000039.tif8157 In some embodiments, each target t, c t,kIt generally increases monotonically and then dips downwards as the value of k increases (each additional peptide increases coverage but decreases return).

[0145] In step 2, the vaccine limit performance is calculated and weighted according to the occurrence rate weighting of each target. For each target t (individually), the limit coverage m of the k-th peptide added to the vaccine set is calculated. t,k Calculate. TIFF0007829242000040.tif15155 In some embodiments, m of each target t t,k This should be a monotonically decreasing function of k (as determined by step 1 above).

[0146] TIFF0007829242000041.tif13159 The weighted limiting population coverage yields the effective limiting coverage of the k-th peptide of the vaccine, weighted by the target's appearance rate in the presentation (by multiplying the target's probability in the presentation by the weighting).

[0147] TIFF0007829242000042.tif62158

[0148] In step 4, a vaccine with the desired performance is selected. The final vaccine size k may vary based on the specific population coverage target of the vaccine. The marginal weighted coverage value M of the combination vaccine. k The cumulative sum over k peptides is used to determine the overall effective (target-weighted) population coverage of a combination vaccine containing k peptides, Σ j≦K M k This is obtained by considering both the probability / weighting of the target in presentation and the expected population coverage of the peptide based on HLA presentation.

[0149] TIFF0007829242000043.tif31159

[0150] Therefore, a combination vaccine having k peptides is optimal for each individual (C t,k It is a combination of ) peptide vaccines.

[0151] MHC class I peptide sequence of RAS vaccine In some embodiments, the peptide vaccine (single-target vaccine or combined multi-target vaccine) contains approximately 5, 10, or 20 MHC class I peptides, each consisting of 8 or more amino acids. In some embodiments, the MHC class I peptide vaccine is intended to target one or more of the KRAS G12D, G12V, and G12R targets. In some embodiments, the amino acid sequence of the first peptide in the five-peptide combination vaccine includes SEQ ID NO: 1.GADGVGKSM (SEQ ID NO: 1). In some embodiments, the amino acid sequence of the second peptide in the five-peptide combination vaccine includes SEQ ID NO: 2.LMVVGADGV (SEQ ID NO: 2). In some embodiments, the amino acid sequence of the third peptide in the five-peptide combination vaccine includes SEQ ID NO: 3.GAVGVGKSL (SEQ ID NO: 3). In some embodiments, the amino acid sequence of the fourth peptide in the five-peptide combination vaccine includes SEQ ID NO: 4.LMVVGAVGV (SEQ ID NO: 4). In some embodiments, the amino acid sequence of the fifth peptide in the five-peptide combination vaccine includes SEQ ID NO: 5.VTGARGVGK (SEQ ID NO: 5). An exemplary combination vaccine having five peptides (SEQ ID NOs: 1 to 5) against KRAS G12D, G12V, and G12R targets is predicted to have a weighted population coverage of 0.3620.

[0152] In some embodiments, one of the peptides (peptides 1-5) in the five-peptide vaccine contains an amino acid sequence that is 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, or 99% identical to SEQ ID NO: 1, SEQ ID NO: 2, SEQ ID NO: 3, SEQ ID NO: 4, or SEQ ID NO: 5.

[0153] In some embodiments, the amino acid sequences of peptides 1-5 in the 10-peptide combination vaccine include SEQ ID NO: 1, SEQ ID NO: 2, SEQ ID NO: 3, SEQ ID NO: 4, and SEQ ID NO: 5. In some embodiments, the amino acid sequence of the 6th peptide in the 10-peptide combination vaccine includes SEQ ID NO: 6.VMGAVGVGK (SEQ ID NO: 6). In some embodiments, the amino acid sequence of the 7th peptide in the 10-peptide combination vaccine includes SEQ ID NO: 7.VVGAVGVGK (SEQ ID NO: 7). In some embodiments, the amino acid sequence of the 8th peptide in the 10-peptide combination vaccine includes SEQ ID NO: 8.GARGVGKSY (SEQ ID NO: 8). In some embodiments, the amino acid sequence of the 9th peptide in the 10-peptide combination vaccine includes SEQ ID NO: 9.GPRGVGKSA (SEQ ID NO: 9). In some embodiments, the amino acid sequence of the 10th peptide in the 10-peptide combination vaccine includes SEQ ID NO: 10.LMVVGARGV (SEQ ID NO: 10). An exemplary combination vaccine containing 10 peptides (SEQ ID NOs: 1 to 10) targeting KRAS G12D, G12V, and G12R is predicted to have a weighted population coverage of 0.4374.

[0154] In some embodiments, one of the peptides (peptides 1-10) in the 10-peptide vaccine contains an amino acid sequence that is 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, or 99% identical to SEQ ID NO: 1, SEQ ID NO: 2, SEQ ID NO: 3, SEQ ID NO: 4, SEQ ID NO: 5, SEQ ID NO: 6, SEQ ID NO: 7, SEQ ID NO: 8, SEQ ID NO: 9, or SEQ ID NO: 10.

[0155] In some embodiments, the amino acid sequences of peptides 1 to 10 in the 20-peptide combination vaccine include SEQ ID NOs: 1, SEQ ID NOs: 2, SEQ ID NOs: 3, SEQ ID NOs: 4, SEQ ID NOs: 5, SEQ ID NOs: 6, SEQ ID NOs: 7, SEQ ID NOs: 8, SEQ ID NOs: 9, and SEQ ID NOs: 10. In some embodiments, the amino acid sequence of the 11th peptide in the 20-peptide combination vaccine includes SEQ ID NOs: 11.GADGVGKSL (SEQ ID NOs: 11). In some embodiments, the amino acid sequence of the 12th peptide in the 20-peptide combination vaccine includes SEQ ID NOs: 12.GADGVGKSY (SEQ ID NOs: 12). In some embodiments, the amino acid sequence of the 13th peptide in the 20-peptide combination vaccine includes SEQ ID NOs: 13.GYDGVGKSM (SEQ ID NOs: 13). In some embodiments, the amino acid sequence of the 14th peptide in the 20-peptide combination vaccine includes SEQ ID NOs: 14.GPVGVGKSV (SEQ ID NOs: 14). In some embodiments, the amino acid sequence of the 15th peptide in the 20-peptide combination vaccine includes SEQ ID NOs: 15.LTVVGAVGV (SEQ ID NOs: 15). In some embodiments, the amino acid sequence of the 16th peptide in the 20-peptide combination vaccine includes SEQ ID NO: 16.VVGAVGVGR (SEQ ID NO: 16). In some embodiments, the amino acid sequence of the 17th peptide in the 20-peptide combination vaccine includes SEQ ID NO: 17.GARGVGKSM (SEQ ID NO: 17). In some embodiments, the amino acid sequence of the 18th peptide in the 20-peptide combination vaccine includes SEQ ID NO: 18.GPRGVGKSV (SEQ ID NO: 18). In some embodiments, the amino acid sequence of the 19th peptide in the 20-peptide combination vaccine includes SEQ ID NO: 19.LLVVGARGV (SEQ ID NO: 19). In some embodiments, the amino acid sequence of the 20th peptide in the 20-peptide combination vaccine includes SEQ ID NO: 20.VAGARGVGM (SEQ ID NO: 20). An exemplary combination vaccine having 20 peptides (SEQ ID NOs: 1 to 20) against KRAS G12D, G12V, and G12R targets is predicted to have a weighted population coverage of 0.4604.

[0156] In some embodiments, any one of the 20 peptides in the vaccine (peptides 1-20) contains an amino acid sequence that is 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, or 99% identical to SEQ ID NOs. 1, SEQ ID NOs. 2, SEQ ID NOs. 3, SEQ ID NOs. 4, SEQ ID NOs. 5, SEQ ID NOs. 6, SEQ ID NOs. 7, SEQ ID NOs. 8, SEQ ID NOs. 9, and SEQ ID NOs. 10, SEQ ID NOs. 11, SEQ ID NOs. 12, SEQ ID NOs. 13, SEQ ID NOs. 14, SEQ ID NOs. 15, SEQ ID NOs. 16, SEQ ID NOs. 17, SEQ ID NOs. 18, SEQ ID NOs. 19, or SEQ ID NOs. 20.

[0157] Table 1 details the MHC class I peptide sequences described herein, including their respective sequence numbers, the amino acid sequences corresponding to the sequence numbers, the KRAS protein target (with specific mutations), the seed amino acid sequence (i.e., the amino acid sequence of the wild-type KRAS fragment), and the amino acid substitutions of the heterocritical peptide at positions 2 and 9 (if present), as well as notes detailing embodiments in which the peptides may be included in the 5, 10, or 20-peptide combination vaccines described herein. Table 1 also includes additional peptide sequences, including sequence numbers 21-41. In some embodiments, any combination of the peptides listed in Table 1 (sequence numbers 1-41) can be used to create combination peptide vaccines having approximately 2 to approximately 40 peptides. In some embodiments, one of the peptides in the combination vaccine (peptides 1-41; SEQ ID NOs: 1-41) contains an amino acid sequence that is 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, or 99% identical to any of SEQ ID NOs: 1-41.

[0158] [Table 1-1] [Table 1-2] [Table 1-3] [Table 1-4]

[0159] Additional amino acid sequences of MHC class I heterocritical peptides are provided in the sequence listings (SEQ ID NOs: 67-1522). In some embodiments, any combination of the MHC class I peptides disclosed herein (SEQ ID NOs: 1-41, 67-1522, 1524-1536, and 1547-1549) can be used to create combination peptide vaccines having about 2 to about 40 peptides. In some embodiments, one of the peptides in the combination vaccine (peptides 1-41, 67-1522, 1524-1536, and 1547-1549) contains an amino acid sequence that is 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, or 99% identical to any of the sequence numbers 1-41, 67-1522, 1524-1536, or 1547-1549.

[0160] MHC class II peptide sequence of the RAS vaccine In some embodiments, the peptide vaccine (single-target vaccine or combination multi-target vaccine) contains approximately 2 to 40 MHC class II peptides, each peptide consisting of approximately 20 amino acids. In some embodiments, the MHC class II peptide vaccine is intended to target one or more of the KRAS G12D, G12V, G12R, G12C, and G13D targets.

[0161] Table 2 summarizes the MHC class II peptide sequences described herein, including each sequence number, the amino acid sequence corresponding to the sequence number, the amino acid sequence corresponding to the peptide binding core, the KRAS protein target (with specific mutations), the seed amino acid sequence (i.e., the amino acid sequence of the wild-type KRAS fragment), the seed amino acid sequence of the binding core, and the amino acid substitutions of the heterocritical peptide at positions 1, 4, 6, and 9 (if present). Table 2 includes peptide sequences including sequence numbers 42-66 and 42-65 (Table 2) that encode recombinant peptides. In some embodiments, any combination of the peptides listed in Table 2 (sequence numbers 42-66) can be used to create single-target (individual) peptide vaccines or combination peptide vaccines having about 2 to about 40 peptides. In some embodiments, one of the peptides in the combination vaccine (peptides 42-66; SEQ ID NOs. 42-66) contains an amino acid sequence that is 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, or 99% identical to any of SEQ ID NOs. 42-66.

[0162] [Table 2-1] [Table 2-2] [Table 2-3]

[0163] In some embodiments, any combination of the MHC class I and / or MHC class II peptides disclosed herein (SEQ ID NOs: 1-1522 and 1524-1549) can be used to create single-target (individual) peptide vaccines or combination peptide vaccines having about 2 to about 40 peptides. In some embodiments, any one of the peptides in the combination vaccine (peptides 1-1522 and 1524-1549; SEQ ID NOs: 1-1522 and 1524-1549) contains an amino acid sequence that is 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, or 99% identical to either SEQ ID NOs: 1-1522 or 1524-1549.

[0164] RAS mRNA and DNA vaccines In some embodiments, the vaccine peptide is encoded as an mRNA or DNA molecule and administered in vivo for expression, as is known in the art. An example of mRNA-based vaccine delivery is found in Kranz et al. (2016) and U.S. Patent No. 8,637,006, which are incorporated herein by reference. In one embodiment, a construct comprising a five-peptide MHC class I combination pancreatic cancer vaccine (targets: KRAS G12D, G12V, G12R) and a five-peptide MHC class II combination pancreatic cancer vaccine (targets: KRAS G12D, G12V, G12R), optimized by the procedure described herein, contains 10 peptides. The peptides are appended with a secretion signal sequence at the N-terminus, followed by an MHC class I transport signal (MITD) (Kreiter et al., 2008; Sahin et al., 2017; U.S. Patent No. 8,637,006). MITD has been shown to guide antigens to pathways for HLA class I and class II presentation (Kreiter et al., 2008). Here, we combine all peptides of each MHC class into a single construct using the non-immunogenic glycine / serine linker of Sahin et al. (2017), although it is also conceivable to construct individual constructs containing a single peptide having the same secretory and MITD signals, as shown by Kreiter et al. (2008).

[0165] In some embodiments, the amino acid sequence encoded by the mRNA vaccine includes SEQ ID NO: 1523. Underlined amino acids correspond to the signal peptide (or leader) sequence. Bold amino acids correspond to MHC class I (9 amino acid length; 5 peptides) and MHC class II (13-25 amino acid length; 5 peptides) peptide sequences. Italicized amino acids correspond to the transport signal. TIFF0007829242000051.tif42159

[0166] In some embodiments, the vaccine is an mRNA vaccine containing a nucleic acid sequence encoding the amino acid sequence of SEQ ID NO: 1523. In some embodiments, the nucleic acid sequence of the mRNA vaccine encodes an amino acid sequence that is 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, or 99% identical to SEQ ID NO: 1523.

[0167] In some embodiments, the vaccine is a DNA vaccine comprising a nucleic acid sequence encoding the amino acid sequence of SEQ ID NO: 1523. In some embodiments, the nucleic acid sequence of the DNA vaccine encodes an amino acid sequence that is 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, or 99% identical to SEQ ID NO: 1523.

[0168] In some embodiments, an mRNA-coding vaccine peptide is used as the payload for an auto-amplified RNA vaccine. In one embodiment, as described in Geall et al. (2012), one or more structural proteins of an infectious alphavirus particle are replaced with an mRNA sequence encoding the vaccine peptide. This document is incorporated herein by reference. As described in Geall et al. (2012), auto-amplified RNA vaccines can increase the efficiency of antigen production in vivo.

[0169] In some embodiments, one or more MHC class I and / or MHC class II peptides disclosed herein (SEQ ID NOs: 1-1522 and 1524-1549) can be encoded in one or more mRNA or DNA molecules and administered in vivo for expression. In some embodiments, about 2 to about 40 peptide sequences are encoded in one or more mRNA constructs. In some embodiments, about 2 to about 40 peptide sequences are encoded in one or more DNA constructs (i.e., nucleic acids encoding amino acid sequences including one or more of SEQ ID NOs: 1-1522 and 1524-1549). In some embodiments, the amino acid sequence of the mRNA vaccine or the nucleic acid sequence of the DNA vaccine encodes an amino acid sequence that is 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, or 99% identical to any of the sequence numbers 1-1522 or 1524-1549.

[0170] MHC class I peptide sequence of the BCR-ABL vaccine In some embodiments, the peptide vaccine (single-target vaccine or combined multi-target vaccine) contains approximately 1 to 40 MHC class I peptides, each peptide consisting of 8 or more amino acids. In some embodiments, the MHC class I peptide vaccine is intended to target a BCR-ABL gene fusion. In some embodiments, the BCR-ABL gene fusion is selected from the group consisting of b3a2 and b2a2. In some embodiments, the MHC class I peptide vaccine is intended to prevent cancer. In some embodiments, the MHC class I peptide vaccine is intended to treat cancer. In some embodiments, the MHC class I peptide vaccine is intended to prevent chronic myeloid leukemia (CML), acute lymphoblastic leukemia (ALL), acute myeloid leukemia (AML), or invasive ductal carcinoma. In some embodiments, the MHC class I peptide vaccine is intended to treat chronic myeloid leukemia (CML), acute lymphoblastic leukemia (ALL), acute myeloid leukemia (AML), or invasive ductal carcinoma.

[0171] In some embodiments, the amino acid sequence vaccine of the MHC class I peptide vaccine against BCR-ABL includes one or more of SEQ ID NOs. 1550-1594. In some embodiments, any one of the peptides in the BCR-ABL vaccine includes an amino acid sequence that is 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, or 99% identical to SEQ ID NOs. 1550-1594.

[0172] In some embodiments, the amino acid sequence vaccine of the MHC class I peptide vaccine against BCR-ABL contains two or more of SEQ ID NOs. 1550-1594. In some embodiments, any one of the peptides in the BCR-ABL vaccine contains an amino acid sequence that is 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, or 99% identical to SEQ ID NOs. 1550-1594.

[0173] Table 3 details the MHC class I peptide sequences described herein, including their respective sequence numbers, the amino acid sequences corresponding to the sequence numbers, the seed amino acid sequence (i.e., the amino acid sequence of the wild-type BCR-ABL protein fusion fragment), and the amino acid substitutions of the heterocritical peptide at positions 2 and C (carboxyl terminus), if present, as well as notes detailing embodiments in which the peptides may be included in the combination peptide vaccines described herein. Sequence numbers 1550-1582 and 1594 are derived from BCR-ABL b3a2, and sequence numbers 1583-1593 are derived from BCR-ABL b2a2. In some embodiments, any combination of the peptides listed in Table 3 (sequence numbers 1550-1594) can be used to create combination peptide vaccines having about 1 to about 40 peptides. In some embodiments, one of the peptides in the combination vaccine (peptides 1550-1594; and 1550-1594) contains an amino acid sequence that is 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, or 99% identical to any of the sequence numbers 1550-1594.

[0174] In some embodiments, a combination peptide vaccine having approximately 1 to approximately 40 peptides can be created using any combination of the peptides listed in the “b3a2 vaccine” column of Table 3 (SEQ ID NOs. 1550-1582 and SEQ ID NOs. 1594). In some embodiments, any one of these peptides in the combination vaccine contains an amino acid sequence that is 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, or 99% identical to the peptides listed in the “b3a2 vaccine” column of Table 3 (SEQ ID NOs. 1550-1582 and SEQ ID NOs. 1594).

[0175] In some embodiments, a combination peptide vaccine having approximately 1 to approximately 40 peptides can be created using any combination of the peptides listed in the “b2a2 vaccine” column of Table 3 (SEQ ID NOs: 1583–1593). In some embodiments, any one of these peptides in the combination vaccine contains an amino acid sequence that is 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, or 99% identical to the peptides listed in the “b2a2 vaccine” column of Table 3 (SEQ ID NOs: 1583–1593).

[0176] Additional amino acid sequences of MHC class I vaccine peptides are provided in the sequence listings (SEQ ID NOs: 1662-2249). In some embodiments, any combination of the MHC class I peptides disclosed herein (SEQ ID NOs: 1550-1594 and SEQ ID NOs: 1662-2249) can be used to create a combination peptide vaccine having about 1 to about 40 peptides. In some embodiments, one of the peptides in the combination vaccine (SEQ ID NOs: 1550-1594 and SEQ ID NOs: 1662-2249) contains an amino acid sequence that is 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, or 99% identical to either SEQ ID NOs: 1550-1594 or SEQ ID NOs: 1662-2249.

[0177] [Table 3-1] [Table 3-2]

[0178] MHC class II peptide sequence of the BCR-ABL vaccine In some embodiments, the peptide vaccine (single-target vaccine or combined multi-target vaccine) contains approximately 1 to 40 MHC class II peptides, each peptide consisting of approximately 20 amino acids. In some embodiments, the MHC class II peptide vaccine is intended to target a BCR-ABL gene fusion. In some embodiments, the BCR-ABL gene fusion is selected from the group consisting of b3a2 and b2a2. In some embodiments, the MHC class II peptide vaccine is intended to prevent cancer. In some embodiments, the MHC class II peptide vaccine is intended to treat cancer. In some embodiments, the MHC class I peptide vaccine is intended to prevent chronic myeloid leukemia (CML), acute lymphoblastic leukemia (ALL), acute myeloid leukemia (AML), or invasive ductal carcinoma. In some embodiments, the MHC class I peptide vaccine is intended to treat chronic myeloid leukemia (CML), acute lymphoblastic leukemia (ALL), acute myeloid leukemia (AML), or invasive ductal carcinoma.

[0179] In some embodiments, the amino acid sequence vaccine of the MHC class II peptide vaccine against BCR-ABL includes one or more of SEQ ID NOs: 1595-1661. In some embodiments, any one of the peptides in the BCR-ABL vaccine includes an amino acid sequence that is 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, or 99% identical to SEQ ID NOs: 1595-1661.

[0180] In some embodiments, the amino acid sequence vaccine of the MHC class II peptide vaccine against BCR-ABL contains two or more of SEQ ID NOs. 1595-1661. In some embodiments, any one of the peptides in the BCR-ABL vaccine contains an amino acid sequence that is 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, or 99% identical to SEQ ID NOs. 1595-1661.

[0181] Table 4 summarizes the MHC class II peptide sequences described herein, including each sequence number, the amino acid sequence corresponding to the sequence number, the amino acid sequence corresponding to the peptide binding core, the seed amino acid sequence (i.e., the amino acid sequence of the wild-type BCR-ABL protein fusion fragment), the seed amino acid sequence of the binding core, and the amino acid substitutions of the heterocritical peptide at positions 1, 4, 6, and 9 (if any). Table 4 includes peptide sequences including sequence numbers 1595-1661. Sequence numbers 1595-1661 (Table 4) encode recombinant peptides. Sequence numbers 1595-1627 are derived from BCR-ABL b3a2, and sequence numbers 1628-1661 are derived from BCR-ABL b2a2. In some embodiments, any combination of the peptides listed in Table 4 (sequence numbers 1595-1661) can be used to create single-target (individual) peptide vaccines or combination peptide vaccines having about 1 to about 40 peptides. In some embodiments, one of the peptides in the combination vaccine (peptides 1595-1661; SEQ ID NOs. 1595-1661) contains an amino acid sequence that is 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, or 99% identical to any of SEQ ID NOs. 1595-1661.

[0182] In some embodiments, a combination peptide vaccine having approximately 1 to approximately 40 peptides can be created using any combination of the peptides (SEQ ID NOs. 1595-1627) listed in the “b3a2 vaccine” column of Table 4. In some embodiments, any one of these peptides in the combination vaccine contains an amino acid sequence that is 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, or 99% identical to the peptides (SEQ ID NOs. 1595-1627) listed in the “b3a2 vaccine” column of Table 4.

[0183] In some embodiments, a combination peptide vaccine having approximately 1 to approximately 40 peptides can be created using any combination of the peptides (SEQ ID NOs. 1628-1661) listed in the “b2a2 vaccine” column of Table 4. In some embodiments, any one of these peptides in the combination vaccine contains an amino acid sequence that is 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, or 99% identical to the peptides (SEQ ID NOs. 1628-1661) listed in the “b2a2 vaccine” column of Table 4.

[0184] Additional amino acid sequences of MHC class II vaccine peptides are provided in the sequence listings (SEQ ID NOs: 2250-63661). In some embodiments, any combination of the MHC class II peptides disclosed herein (SEQ ID NOs: 1595-1661 and SEQ ID NOs: 2250-63661) can be used to create a combination peptide vaccine having about 1 to about 40 peptides. In some embodiments, one of the peptides in the combination vaccine (SEQ ID NOs: 1595-1661 and SEQ ID NOs: 2250-63661) contains an amino acid sequence that is 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, or 99% identical to either SEQ ID NOs: 1595-1661 or SEQ ID NOs: 2250-63661.

[0185] In some embodiments, any combination of the MHC class I and / or MHC class II peptides disclosed herein (SEQ ID NOs: 1550-63662) can be used to create single-target (individual) peptide vaccines or combination peptide vaccines having about 2 to about 40 peptides. In some embodiments, any one of the peptides in the combination vaccine (peptides 1550-63662; SEQ ID NOs: 1550-63662) contains an amino acid sequence that is 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, or 99% identical to any of SEQ ID NOs: 1550-63662.

[0186] [Table 4-1] [Table 4-2] [Table 4-3] [Table 4-4] [Table 4-5]

[0187] BCR-ABL mRNA and DNA vaccine In one embodiment, a construct comprising a 10-peptide MHC class I BCR-ABL b3a2 vaccine and a 10-peptide MHC class II BCR-ABL b3a2 vaccine, optimized by the procedure described herein, comprises 20 peptides. Each peptide is appended with a secretion signal sequence at its N-terminus, followed by an MHC class I transport signal (MITD) (Kreiter et al., 2008; Sahin et al., 2017; U.S. Patent No. 8,637,006; these documents are incorporated herein by reference in their entirety). The MITD has been shown to guide antigens to pathways for HLA class I and class II presentation (Kreiter et al., 2008). Here, the inventors combine all peptides of each MHC class into a single construct using the non-immunogenic glycine / serine linker of Sahin et al. (2017), but it is also conceivable to construct individual constructs containing a single peptide having the same secretory and MITD signals, as shown by Kreiter et al. (2008).

[0188] In some embodiments, the amino acid sequence encoded by the mRNA vaccine includes SEQ ID NO: 63662. Underlined amino acids correspond to the signal peptide (or leader) sequence. Bold amino acids correspond to MHC class I (8–11 amino acid length; 10 peptides) and MHC class II (13–25 amino acid length; 10 peptides) peptide sequences. Italicized amino acids correspond to the transport signal. In alternative embodiments, any number and variant peptide sequences disclosed herein may be included in the mRNA vaccine, which includes the signal peptide sequence and transport signal shown in SEQ ID NO: 63662 below. TIFF0007829242000059.tif76157

[0189] In some embodiments, the vaccine is an mRNA vaccine containing a nucleic acid sequence encoding the amino acid sequence of SEQ ID NO: 63662. In some embodiments, the nucleic acid sequence of the mRNA vaccine encodes an amino acid sequence that is 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, or 99% identical to SEQ ID NO: 63662.

[0190] In some embodiments, the vaccine is a DNA vaccine comprising a nucleic acid sequence encoding the amino acid sequence of sequence number 63662. In some embodiments, the nucleic acid sequence of the DNA vaccine encodes an amino acid sequence that is 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, or 99% identical to sequence number 63662.

[0191] In some embodiments, one or more MHC class I and / or MHC class II peptides disclosed herein (SEQ ID NOs. 1550–63662) can be encoded in one or more mRNA or DNA molecules and administered in vivo for expression. In some embodiments, about 2 to about 40 peptide sequences are encoded in one or more mRNA constructs. In some embodiments, about 2 to about 40 peptide sequences are encoded in one or more DNA constructs (i.e., nucleic acids encoding amino acid sequences including one or more of SEQ ID NOs. 1550–63662). In some embodiments, the amino acid sequence of the mRNA vaccine or the nucleic acid sequence of the DNA vaccine encodes an amino acid sequence that is 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, or 99% identical to any of SEQ ID NOs. 1550–63662.

[0192] Non-limiting embodiments of this subject In one embodiment, the present invention provides nucleic acid sequences encoding at least two amino acid sequences selected from the group consisting of SEQ ID NOs: 1, 2, 3, 4, 5, 6, 8, 9, 10, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, and 41.

[0193] In some embodiments, the nucleic acid sequence is an immunogenic composition. In some embodiments, the nucleic acid sequence is administered in vivo in an expression construct. In some embodiments, in vivo administration of the nucleic acid sequence is configured to produce at least one peptide presented by an HLA class I molecule. In some embodiments, the at least one peptide is a modified or unmodified fragment of a mutant KRAS protein. In some embodiments, the mutant KRAS protein is selected from the group consisting of KRAS G12D, KRAS G12V, KRAS G12R, KRAS G12C, and KRAS G13D. In some embodiments, the nucleic acid sequence is administered to a subject in an effective dose to prevent cancer. In some embodiments, the nucleic acid sequence is administered to a subject in an effective dose to treat cancer.

[0194] In another embodiment, the present invention provides an immunogenic peptide composition comprising at least two peptides selected from the group consisting of SEQ ID NOs: 1, 2, 3, 4, 5, 6, 8, 9, 10, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, and 41.

[0195] In some embodiments, at least one of at least two peptides is presented in the subject by an HLA class I molecule. In some embodiments, at least one of at least two peptides is a modified or unmodified fragment of a mutant KRAS protein. In some embodiments, the mutant KRAS protein is selected from the group consisting of KRAS G12D, KRAS G12V, KRAS G12R, KRAS G12C, and KRAS G13D. In some embodiments, the immunogenic peptide composition is administered to the subject in an effective amount to prevent cancer. In some embodiments, the immunogenic peptide composition is administered to the subject in an effective amount to treat cancer. In some embodiments, the immunogenic peptide composition comprises at least three peptides selected from the group consisting of SEQ ID NOs: 1, 2, 3, 4, 5, 6, 8, 9, 10, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, and 41.

[0196] In another embodiment, the present invention provides a nucleic acid sequence encoding at least one amino acid sequence selected from the group consisting of SEQ ID NOs: 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, and 65.

[0197] In some embodiments, the nucleic acid sequence is an immunogenic composition. In some embodiments, the nucleic acid sequence is administered in vivo in an expression construct. In some embodiments, in vivo administration of the nucleic acid sequence is configured to produce at least one peptide presented by an HLA class II molecule. In some embodiments, the at least one peptide is a modified fragment of a mutant KRAS protein. In some embodiments, the mutant KRAS protein is selected from the group consisting of KRAS G12D, KRAS G12V, KRAS G12R, KRAS G12C, and KRAS G13D. In some embodiments, the nucleic acid sequence is administered to a subject in an effective dose to prevent cancer. In some embodiments, the nucleic acid sequence is administered to a subject in an effective dose to treat cancer.

[0198] In another embodiment, the present invention provides an immunogenic peptide composition comprising at least one peptide selected from the group consisting of SEQ ID NOs: 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, and 65.

[0199] In some embodiments, at least one peptide in the immunogenic peptide composition is presented by an HLA class II molecule. In some embodiments, at least one peptide in the immunogenic peptide composition is a modified or unmodified fragment of a mutant KRAS protein. In some embodiments, the mutant KRAS protein is selected from the group consisting of KRAS G12D, KRAS G12V, KRAS G12R, KRAS G12C, and KRAS G13D. In some embodiments, the immunogenic peptide composition is administered to a subject in an effective amount to prevent cancer. In some embodiments, the immunogenic peptide composition is administered to a subject in an effective amount to treat cancer. In some embodiments, the immunogenic peptide composition comprises at least two peptides selected from the group consisting of SEQ ID NOs: 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, and 65.

[0200] In another embodiment, the present invention relates to a method for forming an immunogenic peptide composition, comprising the steps of: producing a first peptide set by selecting a plurality of unmodified peptide sequences using a processor, wherein at least one of the plurality of unmodified peptide sequences is associated with a tumor neoantigen or autoprotein; determining a plurality of peptide-HLA binding scores for each peptide sequence in the first peptide set; determining whether each peptide sequence in the first peptide set has a peptide-HLA binding score that satisfies a first threshold with respect to at least three HLA alleles; producing a second peptide set comprising the first peptide set and a plurality of modified peptide sequences, wherein each modified peptide sequence in the plurality of modified peptide sequences comprises a substitution of at least one amino acid residue in a peptide sequence in the first peptide set; and determining whether each modified peptide sequence in the second peptide set satisfies a second threshold with respect to at least three HLA alleles. The present invention provides a method comprising: a step of determining whether or not a peptide has a peptide-HLA binding score, wherein a second threshold is more restrictive than a first threshold; and a step of creating a third peptide set by selecting a subset of a second peptide set, wherein the selection includes calculating population coverage, the calculation of population coverage including excluding the peptide-HLA binding score of the modified peptide sequence with respect to the first HLA allele if the peptide-HLA binding score of the unmodified peptide sequence associated with the modified peptide sequence does not meet the first threshold with respect to the first HLA allele, and the selected subset has population coverage exceeding the third threshold; performing an experimental assay to obtain a peptide-HLA immunogenicity metric for at least one peptide sequence in the third peptide set; and forming an immunogenic peptide composition comprising at least one peptide sequence in the third peptide set on which the experimental assay was performed.

[0201] In some embodiments, selecting multiple unmodified peptide sequences to create a first peptide set involves sliding a window of size n across amino acid sequences encoding tumor neoantigens or autoproteins, where n is approximately 8 to 25 amino acid lengths, and n is the length of each peptide sequence in the multiple unmodified peptide sequences in the first peptide set. In some embodiments, each peptide sequence in the first peptide set binds to an HLA class I molecule or an HLA class II molecule. In some embodiments, the method further includes filtering the first peptide set to exclude peptide sequences having a predictive binding core that includes the target residue at the anchor position. In some embodiments, the method further includes substituting at least one amino acid residue in each peptide sequence in the first peptide set, where at least one amino acid residue is present at the anchor position for at least one peptide sequence in the first peptide set. In some embodiments, a first threshold is a binding affinity of less than approximately 1000 nM. In some embodiments, a second threshold is a binding affinity of less than approximately 500 nM. In some embodiments, population coverage is calculated based on the frequency of HLA haplotypes in the human population. In some embodiments, population coverage is calculated based on the frequency of at least three HLA alleles in the human population. In some embodiments, multiple unmodified peptide sequences are derived from tumor neoantigens or autoproteins present in the subject. In some embodiments, a third threshold is a proportion of the human population of approximately 0.7 to approximately 0.8. In some embodiments, the tumor neoantigen or autoprotein is associated with cancer, and the cancer is selected from the group consisting of pancreatic, colon, rectal, kidney, bronchus, lung, uterus, cervix, bladder, liver, and stomach.

[0202] In another aspect, the present invention relates to a method for forming an immunogenic peptide composition, comprising the steps of: producing a first peptide set by selecting a plurality of unmodified peptide sequences using a processor, wherein the plurality of unmodified peptide sequences are associated with tumor neoantigens or autoproteins; determining a plurality of peptide-HLA immunogenicity metrics for each peptide sequence in the first peptide set; determining whether each peptide sequence in the first peptide set has a peptide-HLA immunogenicity metric that satisfies a first threshold with respect to at least three HLA alleles; producing a second peptide set comprising the first peptide set and a plurality of modified peptide sequences, wherein each modified peptide sequence in the plurality of modified peptide sequences comprises a substitution of at least one amino acid residue in a peptide sequence in the first peptide set; and determining whether each modified peptide sequence in the second peptide set has a peptide-HLA immunogenicity metric that satisfies a second threshold with respect to at least three HLA alleles. The present invention provides a method comprising: a step of determining a second threshold, wherein the second threshold is more constrained than the first threshold; and a step of creating a third peptide set by selecting a subset of a second peptide set, the selection comprising calculating predicted vaccine performance, wherein the calculation of predicted vaccine performance comprises excluding the peptide-HLA immunogenicity metric of the modified peptide sequence with respect to the first HLA allele if the peptide-HLA immunogenicity metric of the unmodified peptide sequence associated with the modified peptide sequence does not meet the first threshold with respect to the first HLA allele, and the predicted vaccine performance is a function of the unexcluded peptide-HLA immunogenicity metric of each peptide sequence in the third peptide set; performing an experimental assay to obtain a peptide-HLA immunogenicity metric for at least one peptide sequence in the third peptide set; and forming an immunogenic peptide composition comprising at least one peptide sequence in the third peptide set from which the experimental assay has been performed.

[0203] In some embodiments, selecting multiple unmodified peptide sequences to create a first peptide set involves sliding a window of size n across amino acid sequences encoding tumor neoantigens or autoproteins, where n is approximately 8 to 25 amino acid lengths, and n is the length of each peptide sequence in the multiple unmodified peptide sequences in the first peptide set. In some embodiments, each peptide sequence in the first peptide set binds to an HLA class I molecule or an HLA class II molecule. In some embodiments, the method further includes filtering the first peptide set to exclude peptide sequences having a predictive binding core that includes the target residue at the anchor position. In some embodiments, the method further includes substituting at least one amino acid residue in each peptide sequence in the first peptide set, where at least one amino acid residue is present at the anchor position for at least one peptide sequence in the first peptide set. In some embodiments, the first threshold is a binding affinity of less than approximately 1000 nM.

[0204] In another aspect, the present invention relates to a method for forming an immunogenic peptide composition, comprising the steps of: producing a first peptide set by selecting a plurality of unmodified peptide sequences using a processor, wherein each peptide sequence of the plurality of unmodified peptide sequences is associated with a tumor neoantigen or autoprotein; determining a plurality of peptide-HLA binding scores for each peptide sequence in the first peptide set; determining whether each peptide sequence in the first peptide set has a peptide-HLA binding score that satisfies a first threshold with respect to at least three HLA alleles; producing a second peptide set comprising the first peptide set and a plurality of modified peptide sequences, wherein each modified peptide sequence of the plurality of modified peptide sequences comprises a substitution of at least one amino acid residue in a peptide sequence in the first peptide set; and determining whether each modified peptide sequence in the second peptide set has a peptide-HLA binding score that satisfies a second threshold with respect to at least three HLA alleles. The present invention provides a method comprising: a step of determining whether or not a second threshold is more restrictive than a first threshold; a step of creating a third peptide set by selecting a subset of a second peptide set, the selection comprising calculating predicted vaccine performance, the calculation of predicted vaccine performance comprising excluding the peptide-HLA binding score of a modified peptide sequence with respect to a first HLA allele if the peptide-HLA binding score of an unmodified peptide sequence associated with a modified peptide sequence does not meet a first threshold with respect to a first HLA allele, and the predicted vaccine performance is a function of the unexcluded peptide-HLA binding scores of each peptide sequence in the third peptide set; performing an experimental assay to obtain a peptide-HLA immunogenicity metric for at least one peptide sequence in the third peptide set; and forming an immunogenic peptide composition comprising at least one peptide sequence in the third peptide set on which the experimental assay has been performed.

[0205] In some embodiments, the second threshold is based on data obtained from one or more experimental assays. In some embodiments, the performance of the predicted vaccine is further a function of the peptide-HLA immunogenicity metric of at least one modified peptide sequence among a plurality of modified peptide sequences bound to the second HLA allele among a plurality of HLA alleles, where the first peptide sequence in the first peptide set is predicted to bind to the second HLA allele of at least three HLA alleles by a first binding core, the first binding core is the binding core of the first peptide sequence, the first binding core is identical to the second binding core, the first and second binding cores contain amino acid positions within the peptide sequence, and the second binding core is the binding core of at least one modified peptide sequence among a plurality of modified peptide sequences bound to the second HLA allele.

[0206] In another aspect, the present invention relates to a method for forming an immunogenic peptide composition, comprising the steps of: producing a first peptide set by selecting a plurality of unmodified peptide sequences using a processor, wherein at least one of the plurality of unmodified peptide sequences is associated with a tumor neoantigen or autoprotein; determining a plurality of peptide-HLA binding scores for each peptide sequence in the first peptide set; determining whether each peptide sequence in the first peptide set has a peptide-HLA binding score that satisfies a first threshold with respect to at least three HLA alleles; producing a second peptide set comprising the first peptide set and a plurality of modified peptide sequences, wherein each modified peptide sequence in the plurality of modified peptide sequences comprises a substitution of at least one amino acid residue in a peptide sequence in the first peptide set; and determining whether each modified peptide sequence in the second peptide set satisfies a first threshold with respect to at least three HLA alleles. The present invention provides a method comprising: a step of determining whether a peptide has a peptide-HLA binding score that satisfies a second threshold, wherein the second threshold is more restrictive than a first threshold; a step of creating a third peptide set by selecting a subset of a second peptide set, wherein the selection includes calculating predicted vaccine performance based on the target HLA type, and the calculation of predicted vaccine performance includes excluding the peptide-HLA binding score of the modified peptide sequence with respect to the first HLA allele if the peptide-HLA binding score of the unmodified peptide sequence associated with the modified peptide sequence does not satisfy the first threshold with respect to the first HLA allele; a step of performing an experimental assay to obtain a peptide-HLA immunogenicity metric for at least one peptide sequence in the third peptide set; and a step of forming an immunogenic peptide composition comprising at least one peptide sequence in the third peptide set on which the experimental assay has been performed.

[0207] In some embodiments, each peptide sequence in the first peptide set binds to an HLA class I molecule or an HLA class II molecule. In some embodiments, multiple unmodified peptide sequences are derived from tumor neoantigens or autoproteins present in the subject. In some embodiments, at least three HLA alleles are present in the subject's HLA type. In some embodiments, multiple unmodified peptide sequences are derived from tumor neoantigens or autoproteins present in the subject. In some embodiments, multiple unmodified peptide sequences are derived from tumor neoantigens or autoproteins present in the subject. In some embodiments, the immunogenic peptide composition includes a nucleic acid sequence encoding the amino acid sequence of at least one peptide sequence in the third peptide set. In some embodiments, the immunogenic peptide composition includes a nucleic acid sequence encoding the amino acid sequence of at least one peptide sequence in the third peptide set.

[0208] In another aspect, the present invention provides a composition comprising nucleic acid sequences encoding at least two amino acid sequences selected from the group consisting of SEQ ID NOs: 1550 to 1593.

[0209] In some embodiments, the composition is immunogenic. In some embodiments, the nucleic acid sequence is administered in vivo in a construct for expression. In some embodiments, in vivo administration of the nucleic acid sequence is configured to produce at least one peptide presented by an HLA class I molecule. In some embodiments, the at least one peptide is a modified or unmodified fragment of a BCL-ABL gene fusion. In some embodiments, the BCR-ABL gene fusion is b3a2 or b2a2. In some embodiments, the nucleic acid sequence is administered to a subject in an effective dose to prevent cancer. In some embodiments, the nucleic acid sequence is administered to a subject in an effective dose to treat cancer.

[0210] In another embodiment, the present invention provides a composition comprising at least two aminopeptides selected from the group consisting of SEQ ID NOs: 1550 to 1593.

[0211] In some embodiments, at least one of at least two peptides is presented by an HLA class I molecule in the subject. In some embodiments, at least one peptide in the immunogenic peptide composition is a modified or unmodified fragment of a BCL-ABL gene fusion. In some embodiments, the BCR-ABL gene fusion is b3a2 or b2a2. In some embodiments, the immunogenic peptide composition is administered to the subject in an effective amount to prevent cancer. In some embodiments, the immunogenic peptide composition is administered to the subject in an effective amount to treat cancer. In another embodiment, the present invention provides a composition comprising at least two peptides selected from the group consisting of SEQ ID NOs: 1550-1593.

[0212] In another aspect, the present invention provides a nucleic acid sequence encoding at least one amino acid sequence selected from the group consisting of SEQ ID NOs: 1595 to 1661.

[0213] In some embodiments, the composition is immunogenic. In some embodiments, the nucleic acid sequence is administered in vivo in a construct for expression. In some embodiments, in vivo administration of the nucleic acid sequence is configured to produce at least one peptide presented by an HLA class II molecule. In some embodiments, at least one amino acid sequence is derived from a modified fragment of a BCL-ABL gene fusion. In some embodiments, the BCR-ABL gene fusion is b3a2 or b2a2. In some embodiments, the nucleic acid sequence is administered to a subject in an effective dose to prevent cancer. In some embodiments, the nucleic acid sequence is administered to a subject in an effective dose to treat cancer.

[0214] In another aspect, the present invention provides a nucleic acid sequence encoding at least one amino acid sequence selected from the group consisting of SEQ ID NOs: 1595 to 1661.

[0215] In some embodiments, at least one peptide is presented by an HLA class II molecule in the subject. In some embodiments, at least one peptide in the immunogenic peptide composition is a modified or unmodified fragment of a BCL-ABL gene fusion. In some embodiments, the BCR-ABL gene fusion is b3a2 or b2a2. In some embodiments, the immunogenic peptide composition is administered to the subject in an effective dose to prevent cancer. In some embodiments, the immunogenic peptide composition is administered to the subject in an effective dose to treat cancer. In some embodiments, the immunogenic peptide composition comprises at least two peptides selected from the group consisting of SEQ ID NOs: 1595-1661.

[0216] In another aspect, the present invention relates to a method for forming an immunogenic peptide composition, comprising the steps of: producing a first peptide set by selecting a plurality of unmodified peptide sequences using a processor, wherein at least one of the plurality of unmodified peptide sequences is associated with a tumor neoantigen, a pathogen proteome, or an autoprotein; determining a plurality of peptide-HLA binding scores for each peptide sequence in the first peptide set; determining whether each peptide sequence in the first peptide set has a peptide-HLA binding score that satisfies a first threshold with respect to at least three HLA alleles; and producing a second peptide set comprising the first peptide set and a plurality of modified peptide sequences, wherein each modified peptide sequence in the plurality of modified peptide sequences comprises a substitution of at least one amino acid residue in a peptide sequence in the first peptide set; and each of the second peptide sets The present invention provides a method comprising the steps of: determining multiple peptide-HLA binding scores for each peptide sequence; creating a third peptide set by selecting a subset of a second peptide set, wherein the selection includes calculating population coverage, and the calculation of population coverage includes excluding the peptide-HLA binding score for the first HLA allele of the modified peptide sequence if the peptide-HLA binding score of the unmodified peptide sequence associated with the modified peptide sequence does not meet a first threshold with respect to the first HLA allele, and the selected subset has population coverage exceeding the third threshold; performing an experimental assay to obtain a peptide-HLA immunogenicity metric for at least one peptide sequence in the third peptide set; and forming an immunogenic peptide composition containing at least one peptide sequence from the third peptide set on which the experimental assay was performed.

[0217] In some embodiments, selecting multiple unmodified peptide sequences to create a first peptide set involves sliding a window of size n over at least a portion of amino acid sequences encoding tumor neoantigens, pathogen proteomes, or autoproteins, where n is approximately 8 to 25 amino acid lengths, and n is the length of each peptide sequence in the multiple unmodified peptide sequences of the first peptide set. In some embodiments, each peptide sequence in the first peptide set binds to an HLA class I molecule or an HLA class II molecule. In some embodiments, the first peptide set is further filtered to exclude peptide sequences having a predictive binding core that includes the target amino acid residue at the anchor position. In some embodiments, at least one amino acid residue is substituted in each peptide sequence of the first peptide set, so that for at least one peptide sequence in the first peptide set, at least one amino acid residue is present at the anchor position. In some embodiments, the first threshold is a binding affinity of less than approximately 1000 nM. In some embodiments, population coverage is calculated with respect to at least three HLA alleles. In some embodiments, population coverage is calculated based on the frequency of HLA haplotypes in the human population. In some embodiments, population coverage is calculated based on the frequency of at least three HLA alleles in the human population. In some embodiments, the multiple unmodified peptide sequences are derived from tumor neoantigens, pathogen proteomes, or autoproteins present in the subject. In some embodiments, the second threshold is a proportion of the human population of about 0.7 to about 0.8. In some embodiments, the tumor neoantigen or autoprotein is associated with cancer, which is selected from the group consisting of pancreatic, colon, rectal, kidney, bronchus, lung, uterus, cervix, bladder, liver, and stomach. In some embodiments, the pathogen proteome is associated with pathogen infection in the human subject. In some embodiments, the immunogenic peptide composition includes a nucleic acid sequence encoding the amino acid sequence of at least one peptide sequence of a third peptide set.

[0218] In another aspect, the present invention relates to a method for forming an immunogenic peptide composition, comprising the steps of: producing a first peptide set by selecting a plurality of unmodified peptide sequences using a processor, wherein the plurality of unmodified peptide sequences are associated with tumor neoantigens, pathogen proteomes, or autoproteins; determining a plurality of peptide-HLA immunogenicity metrics for each peptide sequence in the first peptide set; determining whether each peptide sequence in the first peptide set has a peptide-HLA immunogenicity metric that satisfies a threshold with respect to at least three HLA alleles; producing a second peptide set comprising the first peptide set and a plurality of modified peptide sequences, wherein each modified peptide sequence in the plurality of modified peptide sequences comprises a substitution of at least one amino acid residue in a peptide sequence in the first peptide set; and determining whether each peptide sequence in the second peptide set has a plurality of peptide-HLA immunogenicity metrics that satisfy a threshold with respect to at least three HLA alleles. The present invention provides a method comprising the steps of: determining a trick; creating a third peptide set by selecting a subset of a second peptide set, wherein the selection includes calculating predicted vaccine performance, the calculation of predicted vaccine performance including excluding the peptide-HLA immunogenicity metric of the modified peptide sequence with respect to the first HLA allele if the peptide-HLA immunogenicity metric of the unmodified peptide sequence associated with the modified peptide sequence does not meet a threshold with respect to the first HLA allele, and the predicted vaccine performance is a function of the peptide-HLA immunogenicity metric of each peptide sequence in the third peptide set; performing an experimental assay to obtain a peptide-HLA immunogenicity metric for at least one peptide sequence of the third peptide set; and forming an immunogenic peptide composition comprising at least one peptide sequence of the third peptide set on which the experimental assay has been performed.

[0219] In some embodiments, selecting multiple unmodified peptide sequences to create a first peptide set involves sliding a window of size n over at least a portion of amino acid sequences encoding tumor neoantigens, pathogen proteomes, or autoproteins, where n is approximately 8 to 25 amino acid lengths, and n is the length of each peptide sequence in the multiple unmodified peptide sequences of the first peptide set. In some embodiments, each peptide sequence in the first peptide set binds to an HLA class I molecule or an HLA class II molecule. In some embodiments, the first peptide set is further filtered to exclude peptide sequences having a predictive binding core containing the target amino acid residue at the anchor position. In some embodiments, at least one amino acid residue is substituted in each peptide sequence of the first peptide set. In some embodiments, the threshold is a binding affinity of less than approximately 1000 nM. In some embodiments, at least three HLA alleles are present in the target HLA type. In some embodiments, the multiple peptide sequences are derived from tumor neoantigens, pathogen proteomes, or autoproteins present in the target. In some embodiments, the immunogenic peptide composition includes a nucleic acid sequence encoding the amino acid sequence of at least one peptide sequence of a third peptide set.

[0220] In another aspect, the present invention relates to a method for forming an immunogenic peptide composition, comprising the steps of: producing a first peptide set by selecting a plurality of unmodified peptide sequences using a processor, wherein each peptide sequence of the plurality of unmodified peptide sequences is associated with a tumor neoantigen, a pathogen proteome, or an autoprotein; determining a plurality of peptide-HLA binding scores for each peptide sequence in the first peptide set; determining whether each peptide sequence in the first peptide set has a peptide-HLA binding score that satisfies a threshold with respect to at least three HLA alleles; producing a second peptide set comprising the first peptide set and a plurality of modified peptide sequences, wherein each modified peptide sequence of the plurality of modified peptide sequences comprises a substitution of at least one amino acid residue in a peptide sequence in the first peptide set; and determining whether each peptide sequence in the second peptide set has a plurality of peptide- The present invention provides a method comprising the steps of: determining an HLA binding score; creating a third peptide set by selecting a subset of a second peptide set, wherein the selection includes calculating predicted vaccine performance, the calculation of predicted vaccine performance including excluding the peptide-HLA binding score of the modified peptide sequence with respect to the first HLA allele if the peptide-HLA binding score of the unmodified peptide sequence associated with the modified peptide sequence does not meet a threshold for the first HLA allele, and the predicted vaccine performance is a function of the peptide-HLA binding score of each peptide sequence in the third peptide set; performing an experimental assay to obtain a peptide-HLA immunogenicity metric for at least one peptide sequence of the third peptide set; and forming an immunogenic peptide composition comprising at least one peptide sequence of the third peptide set on which the experimental assay was performed.

[0221] In some embodiments, a second threshold is determined from data obtained from one or more experimental assays. In some embodiments, the performance of the predicted vaccine is further a function of the peptide-HLA immunogenicity metric of at least one modified peptide sequence of the second peptide set with respect to the second HLA allele, where the first peptide sequence of the first peptide set is predicted to bind to the second HLA allele by a first binding core, the first binding core is the binding core of the first peptide sequence, the first binding core is identical to the second binding core, and the first and second binding cores each contain an amino acid position within the peptide sequence, the second binding core is the binding core of at least one modified peptide sequence. In some embodiments, the multiple peptide sequences are derived from tumor neoantigens, pathogen proteomes, or autoproteins present in the subject.

[0222] In another aspect, the present invention relates to a method for forming an immunogenic peptide composition, comprising the steps of: producing a first peptide set by selecting a plurality of unmodified peptide sequences using a processor, wherein at least one of the plurality of unmodified peptide sequences is associated with a tumor neoantigen, a pathogen proteome, or an autoprotein; determining a plurality of peptide-HLA binding scores for each peptide sequence in the first peptide set; determining whether each peptide sequence in the first peptide set has a peptide-HLA binding score that satisfies a threshold with respect to at least three HLA alleles; producing a second peptide set comprising the first peptide set and a plurality of modified peptide sequences, wherein each modified peptide sequence in the plurality of modified peptide sequences comprises a substitution of at least one amino acid residue in a peptide sequence in the first peptide set; and producing a second peptide set comprising the first peptide set and a plurality of modified peptide sequences. The present invention provides a method comprising the steps of: determining multiple peptide-HLA binding scores for each peptide sequence in a peptide set; creating a third peptide set by selecting a subset of a second peptide set, wherein the selection includes calculating predicted vaccine performance based on the target HLA type, and the calculation of predicted vaccine performance includes excluding the peptide-HLA binding score of the modified peptide sequence for the first HLA allele if the peptide-HLA binding score of the unmodified peptide sequence associated with the modified peptide sequence does not meet a threshold with respect to the first HLA allele; performing an experimental assay to obtain a peptide-HLA immunogenicity metric for at least one peptide sequence of the third peptide set; and forming an immunogenic peptide composition containing at least one peptide sequence of the third peptide set on which the experimental assay was performed.

[0223] In some embodiments, each peptide sequence in the first peptide set binds to an HLA class I molecule or an HLA class II molecule. In some embodiments, multiple peptide sequences are derived from tumor neoantigens, pathogen proteomes, or autoproteins present in the subject.

[0224] composition In some embodiments, the peptide vaccine comprises one or more peptides of the present disclosure and is administered in a pharmaceutical composition comprising a pharmaceutically acceptable carrier. In some embodiments, the peptide vaccine comprises the third set of peptides described in the present disclosure. In some embodiments, the pharmaceutical composition is in the form of a spray, aerosol, gel, solution, emulsion, lipid nanoparticles, nanoparticles, or suspension. In some embodiments, the pharmaceutical composition is in the form of a cationic nanoemulsion, an example of which is described in Brito et al. (2014). This document is incorporated herein by reference.

[0225] The composition is preferably administered to a subject together with a pharmaceutically acceptable carrier. Typically, in some embodiments, in some embodiments an appropriate amount of a pharmaceutically acceptable salt that can render the formulation isotonic is used in the formulation.

[0226] In certain embodiments, the peptide is provided as an immunogenic composition comprising any one of the peptides described herein and a pharmaceutically acceptable carrier. In certain embodiments, the immunogenic composition further comprises an adjuvant. In certain embodiments, the peptide is conjugated to other molecules to increase its efficacy, as is known to those skilled in the art. For example, the peptide can be coupled to an antibody that recognizes a cell surface protein of an antigen-presenting cell to enhance the efficacy of the vaccine. One such method for increasing the efficacy of peptide delivery is described in Woodham et al. (2018). In certain embodiments for treating autoimmune disorders, the peptide is delivered using compositions and protocols designed to induce tolerance, as is known in the art. Examples of methods for using peptides for immunotolerance induction are described in Alhadj Ali, et al. (2017) and Gibson, et al. (2015).

[0227] In some embodiments, pharmaceutically acceptable carriers are selected from the group consisting of physiological saline, Ringer's solution, dextrose solution, and combinations thereof. Other suitable pharmaceutically acceptable carriers known in the art are intended. Suitable carriers and their formulations are described in Remington's Pharmaceutical Sciences, 2005, Mack Publishing Co. The pH of the solution is preferably about 5 to about 8, more preferably about 7 to about 7.5. The formulation may also contain lyophilized powder. Further carriers include sustained-release preparations such as a semipermeable matrix of a solid hydrophobic polymer, which may be in the form of a molded article, e.g., a film, liposomes, or microparticles. It will be apparent to those skilled in the art that certain carriers may be more preferred, for example, depending on the administration route and concentration of the peptide to be administered.

[0228] The term "pharmaceutically acceptable carrier," as used herein, means a pharmaceutically acceptable substance, composition, or medium, such as a liquid or solid filler, diluent, excipient, solvent, or encapsulant, that is involved in moving or transporting the drug of the subject from one organ or part of the body to another organ or part of the body. Each carrier is acceptable in the sense that it is compatible with the other components of the formulation and is not harmful to the patient. Examples of substances that can serve as pharmaceutically acceptable carriers include: sugars such as lactose, glucose, and sucrose; starches such as corn starch and potato starch; celluloses and their derivatives such as sodium carboxymethylcellulose, ethylcellulose, and cellulose acetate; tragacanth powder; malt; gelatin; talc; excipients such as cocoa butter and suppository wax; oils such as peanut oil, cottonseed oil, safflower oil, sesame oil, olive oil, corn oil, and soybean oil; glycols such as butylene glycol; polyols such as glycerin, sorbitol, mannitol, and polyethylene glycol; esters such as ethyl oleate and ethyl laurate; agar; buffers such as magnesium hydroxide and aluminum hydroxide; alginic acid; non-pyrogenic water; isotonic saline; Ringer's solution; ethyl alcohol; phosphate buffer solution; and other non-toxic, compatible substances used in pharmaceutical formulations. The term carrier refers to natural or synthetic organic or inorganic components that are combined with active ingredients to facilitate application. Furthermore, the components of the pharmaceutical composition can also be mixed with the compounds of the present invention in such a way that no interactions exist that would substantially impair the desired pharmaceutical efficacy. The composition may also contain additional activators such as isotonic agents, preservatives, surfactants, and divalent cations, preferably zinc.

[0229] The composition may also contain excipients or buffers, reducing agents, bulk proteins, amino acids (e.g., glycine or praline), or carbohydrates, which are activators for stabilizing the peptide composition. Albumin is an example of a bulk protein useful for formulating peptide compositions. Typical carbohydrates useful for formulating peptides include, but are not limited to, sucrose, mannitol, lactose, trehalose, or glucose.

[0230] Surfactants can also be used to prevent soluble and insoluble aggregation and / or precipitation of peptides or proteins contained in the composition. Suitable surfactants include, but are not limited to, sorbitan trioleate, soy lecithin, and oleic acid. In certain cases, a solution aerosol using a solvent such as ethanol is preferred. Therefore, a peptide-containing formulation may further contain a surfactant that can reduce or prevent surface-induced aggregation of the peptide by atomization of the solution when forming an aerosol. Various conventional surfactants can be used, such as polyoxyethylene fatty acid esters and alcohols, as well as polyoxyethylene sorbitol fatty acid esters. The amount is generally in the range of 0.001% to 4% by weight of the formulation. In some embodiments, surfactants used with this disclosure include polyoxyethylene sorbitan monooleate, polysorbate 80, and polysorbate 20. Additional agents known in the art may also be included in the composition.

[0231] In some embodiments, the pharmaceutical composition and dosage form further include one or more compounds that reduce the rate at which the active ingredient would disintegrate or the properties of the composition would change. Examples of so-called stabilizers or preservatives include, but are not limited to, amino acids, antioxidants, pH buffers, or salt buffers. Non-limiting examples of antioxidants include butylated hydroxyanisole (BHA), ascorbic acid and its derivatives, tocopherol and its derivatives, butylated hydroxyanisole, and cysteine. Non-limiting examples of preservatives include parabens such as methyl or propyl p-hydroxybenzoate and benzalkonium chloride. Further non-limiting examples of amino acids include glycine or proline.

[0232] Furthermore, the present invention teaches that by using amino acids containing proline or glycine, with or without divalent cations, liquid solutions containing peptides at or below neutral pH can be stabilized (preventing or minimizing thermally or mechanically induced soluble or insoluble aggregation and / or precipitation of inhibitor proteins), resulting in a clear or nearly clear solution that is stable at room temperature or suitable for drug administration.

[0233] In one embodiment, the composition is a pharmaceutical composition in a single-unit dosage form or a multi-unit dosage form. The pharmaceutical composition in a single-unit dosage form or a multi-unit dosage form of the present invention comprises one or more compositions (e.g., the compound of the present invention, or other prophylactic or therapeutic agents) in a prophylactic or therapeutic amount, typically one or more media, carriers, or excipients, stabilizers, and / or preservatives. Preferably, the media, carriers, excipients, stabilizers, and preservatives are pharmaceutically acceptable.

[0234] In some embodiments, the pharmaceutical compositions and dosage forms include anhydrous pharmaceutical compositions and dosage forms. The anhydrous pharmaceutical compositions and dosage forms of the present invention can be prepared using anhydrous or low-moisture-containing components and low-moisture or low-humidity conditions. Pharmaceutical compositions and dosage forms comprising lactose and at least one active ingredient comprising a primary or secondary amine are preferably anhydrous if substantial contact with moisture and / or humidity is expected during manufacturing, packaging, and / or storage. Anhydrous pharmaceutical compositions should be prepared and stored so as to maintain their anhydrous properties. Therefore, anhydrous compositions are preferably packaged using materials known to prevent exposure to water so that they can be included in a suitable formulation kit. Examples of suitable packaging, but not limited to, include airtight foil, plastic, unit dose containers (e.g., vials), blister packs, and strip packs.

[0235] Suitable media are well known to those skilled in the pharmaceutical field, and non-limiting examples of suitable media include glucose, sucrose, starch, lactose, gelatin, rice, silica gel, glycerol, talc, sodium chloride, dried skim milk, propylene glycol, water, sodium stearate, ethanol, and similar substances well known in the art. Physiological saline solutions and aqueous solutions of dextrose and glycerol can also be used as liquid media. Whether a particular media is suitable for incorporation into a pharmaceutical composition or dosage form depends on various factors well known in the art, including how the dosage form will be administered to the patient and the specific active ingredients in the dosage form, but is not limited to these. Pharmaceutical media may be sterile liquids such as water and oil, including petroleum, animal, plant, or synthetic sources such as peanut oil, soybean oil, mineral oil, and sesame oil.

[0236] Furthermore, the present invention provides that the pharmaceutical composition may be packaged in an airtight container such as an ampoule or sachet, in which the quantity is indicated. In one embodiment, the pharmaceutical composition may be supplied as a sterile lyophilized powder in a delivery device suitable for administration to a patient's lower respiratory tract. The pharmaceutical composition may be provided in a pack or dispenser device which may optionally contain one or more unit dosage forms containing the active ingredient. The pack may include, for example, metal foil or plastic foil such as a blister pack. Instructions for administration may be attached to the pack or dispenser device.

[0237] Methods for preparing such formulations or compositions include the step of associating the compound of the present invention with a carrier and, optionally, one or more auxiliary components. Generally, formulations are prepared by associating the compound of the present invention uniformly and tightly with a liquid carrier, or a fine solid carrier, or both, and then, if necessary, shaping the product.

[0238] Formulations of the present invention suitable for administration may be in the form of powders, granules, or solutions or suspensions in aqueous or non-aqueous liquids, or oil-in-water or water-in-oil liquid emulsions, or elixirs or syrups, or lozenges (using inert bases such as gelatin and glycerin, or sucrose and acacia), and / or mouthwashes, each containing a predetermined amount of the compound of the present invention (e.g., a peptide) as an active ingredient.

[0239] The liquid compositions herein can be used as is in delivery devices or in the preparation of pharmaceutically acceptable formulations containing peptides, which may be prepared, for example, by spray drying. A method for spray-freeze-drying peptides / proteins for drug administration, disclosed in Maa et al., Curr. Pharm. Biotechnol., 2001, 1, 283-302, is incorporated herein. In another embodiment, the liquid solutions herein are freeze-spray-dried, and the spray-dried product is collected as a dispersible peptide-containing powder that is therapeutically effective when administered to an individual.

[0240] The compounds and pharmaceutical compositions of the present invention can be used in combination therapy. That is, the compounds and pharmaceutical compositions can be administered simultaneously with, before, or after one or more other desired therapeutic agents or medical procedures (for example, peptide vaccines can be used in combination therapy with other therapies such as chemotherapy, radiation, pharmaceuticals, and / or other treatments). The specific combination of therapies (therapeutic agents or procedures) to be used in a combination regimen will take into account the suitability of the desired therapeutic agents and / or procedures as well as the desired therapeutic effect to be achieved. It will also be understood that the therapies used can achieve the desired effect against the same disorder (for example, the compounds of the present invention can be administered simultaneously with another therapeutic agent or prophylactic agent).

[0241] The present invention also provides a pharmaceutical pack or kit comprising one or more containers filled with one or more components of the pharmaceutical composition of the present invention. Such containers may, as appropriate, be accompanied by a notice in the form prescribed by a government agency that regulates the manufacture, use, or sale of pharmaceutical or biological products, the notice reflecting the government agency's approval of manufacture, use, or sale for human administration.

[0242] The present invention provides a dosage form containing a peptide suitable for the treatment of cancer or other diseases. The dosage form can be formulated, for example, as a spray, aerosol, nanoparticles, liposomes, or other forms known to those skilled in the art. See, for example, Remington's Pharmaceutical Sciences; Remington: The Science and Practice of Pharmacy supra; Pharmaceutical Dosage Forms and Drug Delivery Systems by Howard C., Ansel et al., Lippincott Williams & Wilkins; 7th edition (Oct. 1, 1999).

[0243] Generally, dosage forms used for the acute treatment of a disease may contain one or more of the active ingredients in greater quantities than dosage forms used for the chronic treatment of the same disease. In addition, prophylactically and therapeutically effective dosage forms can vary depending on the conditions. For example, when the objective is to treat cancer or other diseases, a therapeutically effective dosage form may contain peptides with appropriate immunogenic activity. On the other hand, different effective dosages may contain peptides with appropriate immunogenic activity when the objective is to use the peptides of the present invention as a prophylactic agent (e.g., a vaccine) against cancer or another disease / condition. These and other ways in which specific dosage forms encompassed in the present invention differ from each other will be immediately apparent to those skilled in the art. For example, see Remington's Pharmaceutical Sciences, 2005, Mack Publishing Co.; Remington: The Science and Practice of Pharmacy by Gennaro, Lippincott Williams & Wilkins; 20th edition (2003); Pharmaceutical Dosage Forms and Drug Delivery Systems by Howard C. Ansel et al., Lippincott Williams & Wilkins; 7th edition (Oct. 1, 1999); and Encyclopedia of Pharmaceutical Technology, edited by Swarbrick, J. & JC Boylan, Marcel Dekker, Inc., New York, 1988. These documents are incorporated herein by reference in their entirety.

[0244] Furthermore, the pH of a pharmaceutical composition or dosage form can be adjusted to improve the delivery and / or stability of one or more active ingredients. Similarly, the polarity of the solvent carrier, its ionic strength, or its tonicity can be adjusted to improve delivery. Compounds such as stearates can be added to the pharmaceutical composition or dosage form to advantageously alter the hydrophilicity or lipophilicity of one or more active ingredients to improve delivery. In this regard, stearates can also serve as a lipid medium of the formulation, as an emulsifier or surfactant, and as a delivery enhancer or penetration enhancer. Different salts, hydrates, or solvates of the active ingredient can be used to further adjust the properties of the resulting composition.

[0245] The composition can be formulated with a suitable carrier and adjuvant using techniques for obtaining a composition suitable for immunization. The composition can include, for example, but not limited to, adjuvants such as alum, poly IC, MF-59, squalene-based adjuvants, or liposome-based adjuvants suitable for immunization.

[0246] In some embodiments, the composition and method include any suitable agent or immunomodulation that can modulate the mechanism of host immune tolerance and the release of induced antibodies. In certain embodiments, the immunomodulatory agent is administered in an amount and for a time sufficient to transiently modulate the immune response of the subject, for example, to induce an immune response that includes antibodies against tumor neoantigens [i.e., tumor-specific antigens (TSAs)].

[0247] Expression system In certain aspects, the invention provides culturing a cell line that expresses any one of the peptides of the invention in a culture medium comprising any of the peptides described herein.

[0248] Various expression systems for producing recombinant proteins / peptides are known in the art, including prokaryotic (e.g., bacterial) expression systems, plant expression systems, insect expression systems, yeast expression systems, and mammalian expression systems. Suitable cell lines can be transformed, transduced, or transfected with nucleic acids containing the coding sequence of the peptide of the present invention to produce the molecule of interest. Expression vectors containing such nucleic acid sequences, which may be ligated to at least one regulatory sequence to enable the expression of the nucleotide sequence in the host cell, can be introduced by methods known in the art. Those skilled in the art will understand that the design of expression vectors may depend on factors such as the selection of the host cell to be transfected and / or the type and / or amount of the desired protein to be expressed. Enhancer regions are sequences found upstream or downstream of the promoter region in the non-coding DNA region and are known in the art to be important for optimizing expression. If necessary, a viral origin of replication can be used, for example, when using a prokaryotic host for plasmid DNA introduction. However, in eukaryotes, chromosomal integration is the common mechanism of DNA replication. In stable transfection of mammalian cells, only a small fraction of cells can integrate the transgenic DNA into their genome. The expression vector and transfection method used can be contributing factors to the success of the integration event. To stably amplify and express a desired protein, a vector containing the DNA encoding the target protein is stably integrated into the genome of a eukaryotic cell (e.g., a mammalian cell) to result in stable expression of the transfected gene. To identify and select clones that stably express the gene encoding the target protein, a gene encoding a selectable marker (e.g., antibiotic or drug resistance) can be introduced into host cells along with the target gene. Cells containing the target gene can be identified by drug selection, in which case cells with the selectable marker gene integrated will survive in the presence of the drug. Cells without the selectable marker gene integrated will die.Next, surviving cells can be screened for the production of the desired protein molecule.

[0249] Host cell strains that modulate the expression of the inserted sequence or modify and process nucleic acids in a desired specific manner can also be selected. Such modifications (e.g., glycosylation and other post-translational modifications) and processing (e.g., cleavage) of peptide / protein products may be important for peptide / protein function. Different host cell strains have characteristic and specific mechanisms for post-translational processing and modification of proteins and gene products. Therefore, by selecting an appropriate host system or cell strain, correct modification and processing of the expressed target protein can be ensured. Thus, eukaryotic host cells with cellular mechanisms for correct processing, glycosylation, and phosphorylation of primary transcripts may be used.

[0250] Various culture parameters can be used for host cells during culture. Appropriate culture conditions for mammalian cells are well known in the art (Cleveland WL, et al., J lmmunol Methods, 1983, 56(2): 221-234) or can be determined by those skilled in the art [see, for example, Animal Cell Culture: A Practical Approach 2nd Ed., Rickwood, D. and Hames, BD, eds. (Oxford University Press: New York, 1992)]. Cell culture conditions may vary depending on the type of host cell selected. Commercially available culture media can be used.

[0251] The peptides of the present invention can be purified from any human or non-human cells expressing the polypeptide, including cells transfected with an expression construct expressing the peptides of the present invention. For protein recovery, isolation, and / or purification, the cell culture medium or cell lysate is centrifuged to remove particulate cells and cell debris. The desired polypeptide molecule is isolated or purified from contaminants, which are soluble proteins and polypeptides, by a suitable purification technique. Non-limited methods for protein purification include: size exclusion chromatography; affinity chromatography; ion exchange chromatography; ethanol precipitation; reversed-phase HPLC; chromatography with resins such as silica or cation exchange resins, e.g., DEAE; chromatofocusing; SDS-PAGE; ammonium sulfate precipitation; gel filtration using Sephadex G-75, Sepharose, e.g.; and protein A Sepharose chromatography for removing immunoglobulin contaminants. Other additives, such as protease inhibitors (e.g., PMSF or proteinase K), can be used to inhibit proteolysis during purification. Additionally, purification procedures that allow for the selection of carbohydrates, such as ion-exchange softgel chromatography or HPLC using cation-exchange or anion-exchange resins, can be used to collect more acidic fractions.

[0252] Treatment methods In some embodiments, the subject matter disclosed herein relates to preventive medical treatment initiated after a diagnosis of cancer to prevent disease progression or to cure the disease. In some embodiments, the term “prevent” means providing a subject with medical treatment to prevent the onset of a potential disease (e.g., one or more types of cancer) (e.g., by administering nucleic acids or peptides disclosed herein), but “prevent” does not necessarily mean that 100% of subjects who receive medical treatment will not develop the disease. In one embodiment, the subject matter disclosed herein relates to the prevention of subjects considered to be at risk of cancer or subjects previously diagnosed with cancer (or another disease). In one embodiment, a subject may be administered a peptide vaccine or a pharmaceutical composition thereof as described herein. The present invention intends to use any of the peptides produced by the systems and methods described herein. In one embodiment, the peptide vaccine as described herein may be administered subcutaneously by syringe or by other suitable methods known in the art.

[0253] The compounds, combinations of compounds, or pharmaceutical compositions disclosed herein may be administered to cells, mammals, or humans by any suitable means. Non-limiting examples of methods of administration include, among others: (a) administration by oral route, including administration in capsules, tablets, granules, sprays, syrups, or other such forms; (b) administration by parenteral routes such as intraocular, intranasal, intraauricular, rectal, vaginal, urethral, ​​transmucosal, buccal, or transdermal, including administration as aqueous suspensions or oily preparations, or as infusions, sprays, suppositories, ointments, etc.; (c) administration by injection, including subcutaneous, intraperitoneal, intravenous, intramuscular, intradermal, intraorbital, intracapsular, intraspinal, or intrasternal, including infusion pump delivery; (d) local administration, such as by depot implantation, including direct injection into the renal or cardiac region; (e) local administration, such as which a person skilled in the art would consider appropriate for bringing the compounds or combinations of compounds disclosed herein into contact with living tissue; (f) administration by inhalation, including aerosolized, sprayed, and powdered formulations; and (g) administration by implantation.

[0254] As will be immediately apparent to those skilled in the art, the effective in vivo dose and specific administration method will vary depending on age, weight, and species being treated, as well as the specific use in which the compound or combination of compounds disclosed herein is used. Determining the effective dose level, i.e., the dose level required to achieve the desired result, can be achieved by those skilled in the art using routine pharmacological methods. Typically, human clinical applications of a product begin at lower dose levels and increase until the desired effect is achieved. Alternatively, useful doses and administration routes of the compositions identified herein can be established using established pharmacological methods with acceptable in vitro studies. Effective animal doses in in vivo studies can be converted to appropriate human doses using conversion methods known in the art (see, for example, Nair AB, Jacob S. A simple practice guide for dose conversion between animals and human. Journal of basic and clinical pharmacy. 2016 Mar;7(2):27).

[0255] Prevention methods In some embodiments, peptides prepared using the methods of the present invention can be used as vaccines to promote an immune response against cancer (e.g., against tumor neoantigens). In some embodiments, the present invention provides compositions and methods for inducing an immune response, such as antibodies against tumor neoantigens. In some embodiments, the antibodies are broad-spectrum neutralizing antibodies. In some embodiments, peptides prepared using the methods of the present invention can be used as vaccines to promote an immune response against pathogens. In some embodiments, peptides prepared using the methods of the present invention can be used as agents for the treatment of autoimmune diseases to promote immune tolerance.

[0256] In some embodiments, peptides prepared using the methods of the present invention can be combined with additional vaccine components. In some embodiments, such combination vaccines may be encoded in one or more nucleic acids encoding peptides produced using the methods described herein and additional vaccine components known in the art (e.g., peptides or proteins). In some embodiments, such combination vaccines are produced by adding peptides or proteins encoding additional vaccine components to peptides resulting from the methods described herein with respect to combination formulations and packaging. An example of vaccine component combination is the production of a RAS vaccine, in which one or more nucleic acids encoding vaccine components against KRAS G12D and KRAS G12V are used, and such nucleic acids are packaged into mRNA-LNP or DNA formulations, or the different components are formulated separately as mRNA-LNP or DNA and then combined for packaging or immediately before administration to an individual.

[0257] Embodiments of the present invention are further described in the following sections: [Section 1] A nucleic acid sequence encoding at least two amino acid sequences selected from the group consisting of SEQ ID NOs: 1, 2, 3, 4, 5, 6, 8, 9, 10, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, and 41. [Section 2] The nucleic acid sequence described in item 1 above, which is an immunogenic composition. [Section 3] The nucleic acid sequence described in item 1 above, administered in vivo in a construct for expression. [Section 4] The nucleic acid sequence according to item 3, wherein the in vivo administration of the nucleic acid sequence is configured to produce at least one peptide presented by an HLA class I molecule. [Section 5] The nucleic acid sequence according to item 4, wherein the at least one peptide is a modified or unmodified fragment of a mutant KRAS protein. [Section 6] The mutant KRAS protein is a nucleic acid sequence selected from the group consisting of KRAS G12D, KRAS G12V, KRAS G12R, KRAS G12C, and KRAS G13D, as described in item 5 above. [Section 7] The nucleic acid sequence described in item 3 above, administered to the target in an effective dose for the prevention of cancer. [Section 8] The nucleic acid sequence described in item 3 above, administered to a target in an effective dose for the treatment of cancer. [Section 9] An immunogenic peptide composition comprising at least two peptides selected from the group consisting of SEQ ID NOs: 1, 2, 3, 4, 5, 6, 8, 9, 10, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, and 41. [Section 10] The immunogenic peptide composition according to item 9, wherein at least one of the two peptides is presented by an HLA class I molecule in the subject. [Section 11] The immunogenic peptide composition according to item 9, wherein at least one of the at least two peptides is a modified or unmodified fragment of a mutant KRAS protein. [Section 12] The mutant KRAS protein is selected from the group consisting of KRAS G12D, KRAS G12V, KRAS G12R, KRAS G12C, and KRAS G13D, as described in item 11 above, in the immunogenic peptide composition. [Section 13] An immunogenic peptide composition as described in item 9 above, administered to a target in an effective amount for the prevention of cancer. [Section 14] An immunogenic peptide composition as described in item 9 above, administered to a target in an effective dose for the treatment of cancer. [Section 15] The immunogenic peptide composition according to item 9 above, comprising at least three peptides selected from the group consisting of SEQ ID NOs: 1, 2, 3, 4, 5, 6, 8, 9, 10, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, and 41. [Section 16] A nucleic acid sequence encoding at least one amino acid sequence selected from the group consisting of SEQ ID NOs: 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, and 65. [Section 17] An immunogenic composition comprising the nucleic acid sequence described in item 16 above. [Section 18] The nucleic acid sequence described in item 16 above, administered in vivo in a construct for expression. [Section 19] The nucleic acid sequence according to item 18, wherein the in vivo administration of the nucleic acid sequence is configured to produce at least one peptide presented by an HLA class II molecule. [Section 20] The nucleic acid sequence described in item 16 above, wherein the at least one peptide is a modified fragment of a mutant KRAS protein. [Section 21] The mutant KRAS protein is a nucleic acid sequence selected from the group consisting of KRAS G12D, KRAS G12V, KRAS G12R, KRAS G12C, and KRAS G13D, as described in item 20 above. [Section 22] The nucleic acid sequence described in item 16 above, administered to the target in an effective dose for the prevention of cancer. [Section 23] A nucleic acid sequence as described in item 16 above, administered to a target in an effective dose for the treatment of cancer. [Section 24] An immunogenic peptide composition comprising at least one peptide selected from the group consisting of SEQ ID NOs: 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, and 65. [Section 25] The immunogenic peptide composition according to item 24, wherein at least one peptide in the immunogenic peptide composition is presented by an HLA class II molecule. [Section 26] The immunogenic peptide composition according to item 24, wherein at least one peptide in the immunogenic peptide composition is a modified or unmodified fragment of a mutant KRAS protein. [Section 27] The mutant KRAS protein is selected from the group consisting of KRAS G12D, KRAS G12V, KRAS G12R, KRAS G12C, and KRAS G13D, as described in item 26 above, for the immunogenic peptide composition. [Section 28] An immunogenic peptide composition as described in item 24 above, administered to a target in an effective amount for the prevention of cancer. [Section 29] An immunogenic peptide composition as described in item 24 above, administered to a target in an effective dose for the treatment of cancer. [Section 30] The immunogenic peptide composition according to item 24, comprising at least two peptides selected from the group consisting of SEQ ID NOs: 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, and 65. [Section 31] A method for forming an immunogenic peptide composition, Using the processor, A step of creating a first peptide set by selecting multiple unmodified peptide sequences, wherein at least one of the multiple unmodified peptide sequences is associated with a tumor neoantigen or an autoprotein. A step of determining multiple peptide-HLA binding scores for each peptide sequence in the first peptide set, A step of determining whether each peptide sequence in the first peptide set has a peptide-HLA binding score that satisfies a first threshold with respect to at least three HLA alleles. A step of producing a second peptide set comprising the first peptide set and a plurality of modified peptide sequences, wherein each of the plurality of modified peptide sequences comprises the substitution of at least one amino acid residue of a peptide sequence in the first peptide set. A step of determining whether each modified peptide sequence in the second peptide set has a peptide-HLA binding score that satisfies a second threshold with respect to the at least three HLA alleles, wherein the second threshold is more restrictive than the first threshold, and A step of creating a third peptide set by selecting a subset of the second peptide set, wherein the selection includes calculating population coverage, the calculation of population coverage includes excluding the peptide-HLA binding score of the modified peptide sequence with respect to the first HLA allele if the peptide-HLA binding score of the unmodified peptide sequence associated with the modified peptide sequence does not meet the first threshold, and the selected subset has population coverage exceeding the third threshold. To implement; Perform experimental assays to obtain a peptide-HLA immunogenicity metric for at least one peptide sequence in the third peptide set; and To form an immunogenic peptide composition comprising the at least one peptide sequence in the third peptide set on which the experimental assay was performed. A method that includes this. [Section 32] The method according to item 31, wherein selecting the plurality of unmodified peptide sequences to produce the first peptide set comprises sliding a window of size n over amino acid sequences encoding the tumor neoantigen or the autoprotein, where n is about 8 amino acid lengths to about 25 amino acid lengths, and n is the length of each peptide sequence of the plurality of unmodified peptide sequences in the first peptide set. [Section 33] The method according to item 31, wherein each peptide sequence in the first peptide set is bound to an HLA class I molecule or an HLA class II molecule. [Section 34] The method according to item 31, further comprising filtering the first set of peptides to exclude peptide sequences having a predictive binding core that includes a target residue at an anchor position. [Section 35] The method according to item 31, further comprising substituting at least one amino acid residue in each peptide sequence in the first peptide set, wherein for at least one peptide sequence in the first peptide set, the at least one amino acid residue is located at an anchor position. [Section 36] The method according to item 31, wherein the first threshold is a binding affinity of less than approximately 1000 nM. [Section 37] The population coverage rate is calculated with respect to the at least three HLA alleles, as described in item 31 above. [Section 38] The method according to item 31, wherein the second threshold is a binding affinity of less than approximately 500 nM. [Section 39] The population coverage rate is calculated based on the frequency of HLA haplotypes in the human population, as described in item 31 above. [Section 40] The method according to item 31, wherein the population coverage rate is calculated based on the frequencies of the at least three HLA alleles in the human population. [Section 41] The method according to item 31, wherein the plurality of unmodified peptide sequences are derived from the tumor neoantigen or the autoprotein present in the target. [Section 42] The method according to item 31 above, wherein the third threshold is the proportion of the human population of approximately 0.7 to approximately 0.8. [Section 43] The method according to item 31, wherein the tumor neoantigen or the autoprotein is associated with cancer, and the cancer is selected from the group consisting of pancreas, colon, rectum, kidney, bronchus, lung, uterus, cervix, bladder, liver, and stomach. [Section 44] A method for forming an immunogenic peptide composition, Using the processor, A step of creating a first peptide set by selecting multiple unmodified peptide sequences, wherein the multiple unmodified peptide sequences are associated with tumor neoantigens or autoproteins. A step of determining a plurality of peptide-HLA immunogenicity metrics for each peptide sequence in the first peptide set; a step of determining whether each peptide sequence in the first peptide set has a peptide-HLA immunogenicity metric that satisfies a first threshold with respect to at least three HLA alleles; A step of producing a second peptide set comprising the first peptide set and a plurality of modified peptide sequences, wherein each of the plurality of modified peptide sequences comprises the substitution of at least one amino acid residue of a peptide sequence in the first peptide set. A step of determining whether each modified peptide sequence in the second peptide set has a peptide-HLA immunogenicity metric that satisfies a second threshold with respect to the at least three HLA alleles, wherein the second threshold is more restrictive than the first threshold, and A step of creating a third peptide set by selecting a subset of the second peptide set, wherein the selection includes calculating predicted vaccine performance, the calculation of predicted vaccine performance includes excluding the peptide-HLA immunogenicity metric of the modified peptide sequence with respect to the first HLA allele if the peptide-HLA immunogenicity metric of the unmodified peptide sequence associated with the modified peptide sequence does not meet the first threshold with respect to the first HLA allele, and the predicted vaccine performance is a function of the excluded peptide-HLA immunogenicity metric of each peptide sequence in the third peptide set. To implement; Perform experimental assays to obtain a peptide-HLA immunogenicity metric for at least one peptide sequence in the third peptide set; and To form an immunogenic peptide composition comprising the at least one peptide sequence in the third peptide set on which the experimental assay was performed. A method that includes this. [Section 45] The method according to item 44, wherein selecting the plurality of unmodified peptide sequences to create the first peptide set comprises sliding a window of size n over amino acid sequences encoding the tumor neoantigen or the autoprotein, where n is about 8 amino acid lengths to about 25 amino acid lengths, and n is the length of each peptide sequence of the plurality of unmodified peptide sequences in the first peptide set. [Section 46] The method according to item 44, wherein each peptide sequence in the first peptide set is bound to an HLA class I molecule or an HLA class II molecule. [Section 47] The method according to item 44, further comprising filtering the first set of peptides to exclude peptide sequences having a predictive binding core that includes a target residue at an anchor position. [Section 48] The method according to item 44, further comprising substituting at least one amino acid residue in each peptide sequence in the first peptide set. [Section 49] The method according to item 44, wherein the first threshold is a binding affinity of less than approximately 1000 nM. [Section 50] A method for forming an immunogenic peptide composition, Using the processor, A step of creating a first peptide set by selecting multiple unmodified peptide sequences, wherein each peptide sequence of the multiple unmodified peptide sequences is associated with a tumor neoantigen or an autoprotein. A step of determining a plurality of peptide-HLA binding scores for each peptide sequence in the first peptide set; a step of determining whether each peptide sequence in the first peptide set has a peptide-HLA binding score that satisfies a first threshold with respect to at least three HLA alleles; A step of producing a second peptide set comprising the first peptide set and a plurality of modified peptide sequences, wherein each of the plurality of modified peptide sequences comprises the substitution of at least one amino acid residue of a peptide sequence in the first peptide set. A step of determining whether each modified peptide sequence in the second peptide set has a peptide-HLA binding score that satisfies a second threshold with respect to the at least three HLA alleles, wherein the second threshold is more restrictive than the first threshold, and A step of creating a third peptide set by selecting a subset of the second peptide set, wherein the selection includes calculating predicted vaccine performance, the calculation of predicted vaccine performance includes excluding the peptide-HLA binding score of the modified peptide sequence with respect to the first HLA allele if the peptide-HLA binding score of the unmodified peptide sequence associated with the modified peptide sequence does not meet the first threshold with respect to the first HLA allele, and the predicted vaccine performance is a function of the excluded peptide-HLA binding scores of each peptide sequence in the third peptide set. To implement; Perform experimental assays to obtain a peptide-HLA immunogenicity metric for at least one peptide sequence in the third peptide set; and To form an immunogenic peptide composition comprising the at least one peptide sequence in the third peptide set on which the experimental assay was performed. A method that includes this. [Section 51] The method according to item 50, wherein the second threshold is based on data obtained from one or more experimental assays. [Section 52] The method according to item 50, wherein the performance of the predicted vaccine is further a function of the peptide-HLA immunogenicity metric of at least one modified peptide sequence among the plurality of modified peptide sequences bound to the second HLA allele of the at least three HLA alleles, provided that the first peptide sequence in the first peptide set is predicted to bind to the second HLA allele of the at least three HLA alleles by a first binding core, the first binding core is the binding core of the first peptide sequence, the first binding core is identical to the second binding core, the first and second binding cores include an amino acid position within the peptide sequence, and the second binding core is the binding core of at least one modified peptide sequence among the plurality of modified peptide sequences bound to the second HLA allele. [Section 53] A method for forming an immunogenic peptide composition, Using the processor, A step of creating a first peptide set by selecting multiple unmodified peptide sequences, wherein at least one of the multiple unmodified peptide sequences is associated with a tumor neoantigen or an autoprotein; a step of determining multiple peptide-HLA binding scores for each peptide sequence in the first peptide set; A step of determining whether each peptide sequence in the first peptide set has a peptide-HLA binding score that satisfies a first threshold with respect to at least three HLA alleles. A step of producing a second peptide set comprising the first peptide set and a plurality of modified peptide sequences, wherein each of the plurality of modified peptide sequences comprises the substitution of at least one amino acid residue of a peptide sequence in the first peptide set. A step of determining whether each modified peptide sequence in the second peptide set has a peptide-HLA binding score that satisfies a second threshold with respect to the at least three HLA alleles, wherein the second threshold is more restrictive than the first threshold. A step of creating a third peptide set by selecting a subset of the second peptide set, wherein the selection includes calculating predicted vaccine performance based on the target HLA type, and the calculation of the predicted vaccine performance includes excluding the peptide-HLA binding score of the modified peptide sequence with respect to the first HLA allele if the peptide-HLA binding score of the unmodified peptide sequence associated with the modified peptide sequence does not meet the first threshold with respect to the first HLA allele. To implement; Perform experimental assays to obtain a peptide-HLA immunogenicity metric for at least one peptide sequence in the third peptide set; and To form an immunogenic peptide composition comprising the at least one peptide sequence in the third peptide set on which the experimental assay was performed. A method that includes this. [Section 54] The method according to item 53, wherein each peptide sequence in the first peptide set is bound to an HLA class I molecule or an HLA class II molecule. [Section 55] The method according to item 54, wherein the plurality of unmodified peptide sequences are derived from the tumor neoantigen or the autoprotein present in the target. [Section 56] The method described in item 54 above, wherein the aforementioned at least three HLA alleles are present in the target HLA type. [Section 57] The method according to item 53, wherein the plurality of unmodified peptide sequences are derived from the tumor neoantigen or the autoprotein present in the target. [Section 58] The method according to item 53, wherein the plurality of unmodified peptide sequences are derived from the tumor neoantigen or the autoprotein present in the subject. [Section 59] The method according to item 53, wherein the immunogenic peptide composition comprises a nucleic acid sequence encoding the amino acid sequence of at least one peptide sequence in the third peptide set. [Section 60] The method according to item 53, wherein the immunogenic peptide composition comprises a nucleic acid sequence encoding the amino acid sequence of at least one peptide sequence in the third peptide set. [Section 61] A composition comprising nucleic acid sequences encoding at least two amino acid sequences selected from the group consisting of SEQ ID NOs: 1550 to 1593. [Section 62] The composition described in item 61 above, which is immunogenic. [Section 63] The composition according to item 61, wherein the nucleic acid sequence is administered in vivo in an expression construct. [Section 64] The composition according to item 63, wherein the in vivo administration of the nucleic acid sequence is configured to produce at least one peptide presented by an HLA class I molecule. [Section 65] The composition according to item 64, wherein the at least one peptide is a modified or unmodified fragment of a BCL-ABL gene fusion. [Section 66] The composition according to item 65, wherein the BCR-ABL gene fusion is b3a2 or b2a2. [Section 67] The composition according to item 63, wherein the nucleic acid sequence is administered to a target in an effective amount for the prevention of cancer. [Section 68] The composition according to item 63, wherein the nucleic acid sequence is administered to a subject in an effective amount for the treatment of cancer. [Section 69] An immunogenic peptide composition comprising at least two peptides selected from the group consisting of SEQ ID NOs: 1550 to 1593. [Section 70] The immunogenic peptide composition according to item 69, wherein at least one of the two peptides is presented by an HLA class I molecule in the subject. [Section 71] The immunogenic peptide composition according to item 69, wherein at least one peptide in the immunogenic peptide composition is a modified or unmodified fragment of a BCL-ABL gene fusion. [Section 72] The immunogenic peptide composition according to item 71, wherein the BCR-ABL gene fusion is b3a2 or b2a2. [Section 73] An immunogenic peptide composition as described in item 69 above, administered to a target in an effective amount for the prevention of cancer. [Section 74] An immunogenic peptide composition as described in item 69 above, administered to a target in an effective dose for the treatment of cancer. [Section 75] The immunogenic peptide composition according to item 69 above, comprising at least three peptides selected from the group consisting of SEQ ID NOs: 1550 to 1593. [Section 76] A nucleic acid sequence encoding at least one amino acid sequence selected from the group consisting of SEQ ID NOs: 1595 to 1661. [Section 77] An immunogenic composition comprising the nucleic acid sequence described in item 76 above. [Section 78] The nucleic acid sequence described in item 76 above, administered in vivo in an expression construct. [Section 79] The nucleic acid sequence according to item 78, wherein the in vivo administration of the nucleic acid sequence is configured to produce at least one peptide presented by an HLA class II molecule. [Section 80] The nucleic acid sequence according to item 76, wherein the at least one amino acid sequence is derived from a modified fragment of the BCL-ABL gene fusion. [Section 81] The nucleic acid sequence according to item 80, wherein the BCR-ABL gene fusion is b3a2 or b2a2. [Section 82] A nucleic acid sequence as described in item 76 above, administered to a target in an effective dose for the prevention of cancer. [Section 83] The nucleic acid sequence described in item 76 above, administered to a target in an effective dose for the treatment of cancer. [Section 84] An immunogenic peptide composition comprising at least one peptide selected from the group consisting of SEQ ID NOs: 1595 to 1661. [Section 85] The immunogenic peptide composition according to item 84, wherein the at least one peptide is presented by an HLA class II molecule in the subject. [Section 86] The immunogenic peptide composition according to item 84, wherein at least one peptide in the immunogenic peptide composition is a modified or unmodified fragment of a BCL-ABL gene fusion. [Section 87] The immunogenic peptide composition according to item 86, wherein the BCR-ABL gene fusion is b3a2 or b2a2. [Section 88] An immunogenic peptide composition as described in item 84 above, administered to a target in an effective amount for the prevention of cancer. [Section 89] An immunogenic peptide composition as described in item 84 above, administered to a target in an effective dose for the treatment of cancer. [Section 90] The immunogenic peptide composition according to item 84 above, comprising at least two peptides selected from the group consisting of SEQ ID NOs: 1595 to 1661. [Section 91] A method for forming an immunogenic peptide composition, Using the processor, A step of creating a first peptide set by selecting multiple unmodified peptide sequences, wherein at least one of the multiple unmodified peptide sequences is associated with a tumor neoantigen, a pathogen proteome, or an autoprotein. A step of determining multiple peptide-HLA binding scores for each peptide sequence in the first peptide set, A step of determining whether each peptide sequence in the first peptide set has a peptide-HLA binding score that satisfies a first threshold with respect to at least three HLA alleles. A step of producing a second peptide set comprising the first peptide set and a plurality of modified peptide sequences, wherein each of the plurality of modified peptide sequences comprises the substitution of at least one amino acid residue of a peptide sequence in the first peptide set. A step of determining multiple peptide-HLA binding scores for each peptide sequence in the second peptide set, and A step of creating a third peptide set by selecting a subset of the second peptide set, wherein the selection includes calculating population coverage, the calculation of population coverage includes excluding the peptide-HLA binding score of the modified peptide sequence with respect to the first HLA allele if the peptide-HLA binding score of the unmodified peptide sequence associated with the modified peptide sequence does not meet the first threshold with respect to the first HLA allele, and the selected subset has population coverage exceeding the second threshold. To implement; Perform experimental assays to obtain a peptide-HLA immunogenicity metric for at least one peptide sequence of the third peptide set; and Forming an immunogenic peptide composition comprising at least one peptide sequence of the third peptide set on which an experimental assay was performed. A method that includes this. [Section 92] The method according to item 91, wherein selecting the plurality of unmodified peptide sequences to produce the first peptide set comprises sliding a window of size n over at least a portion of the amino acid sequences encoding the tumor neoantigen, the pathogen proteome, or the self-protein, where n is about 8 amino acid lengths to about 25 amino acid lengths, and n is the length of each peptide sequence of the plurality of unmodified peptide sequences in the first peptide set. [Section 93] The method according to item 91, wherein each peptide sequence of the first peptide set is bound to an HLA class I molecule or an HLA class II molecule. [Section 94] The method according to item 91, further comprising filtering the first set of peptides to exclude peptide sequences having a predictive binding core containing a target amino acid residue at an anchor position. [Section 95] The method according to item 91, further comprising substituting at least one amino acid residue in each peptide sequence of the first peptide set, wherein for at least one peptide sequence of the first peptide set, the at least one amino acid residue is located at an anchor position. [Section 96] The method according to item 91, wherein the first threshold is a binding affinity of less than approximately 1000 nM. [Section 97] The population coverage rate is calculated with respect to the at least three HLA alleles, according to the method described in paragraph 91 above. [Section 98] The population coverage rate is calculated based on the frequency of HLA haplotypes in the human population, as described in paragraph 91 above. [Section 99] The method according to paragraph 91, wherein the population coverage rate is calculated based on the frequencies of the at least three HLA alleles in the human population. [Section 100] The method according to item 91, wherein the plurality of unmodified peptide sequences are derived from the tumor neoantigen, the pathogen proteome, or the autoprotein present in the target. [Section 101] The method according to item 91 above, wherein the second threshold is the proportion of the human population of approximately 0.7 to approximately 0.8. [Section 102] The method according to item 91, wherein the tumor neoantigen or the autoprotein is associated with cancer, and the cancer is selected from the group consisting of the pancreas, colon, rectum, kidney, bronchus, lung, uterus, cervix, bladder, liver, and stomach. [Section 103] The method according to item 91, wherein the pathogen proteome is associated with pathogen infection in human subjects. [Section 104] The method according to item 91, wherein the immunogenic peptide composition comprises a nucleic acid sequence encoding the amino acid sequence of at least one peptide sequence of the third peptide set. [Section 105] A method for forming an immunogenic peptide composition, Using the processor, A step of creating a first peptide set by selecting multiple unmodified peptide sequences, wherein the multiple unmodified peptide sequences are associated with tumor neoantigens, pathogen proteomes, or autoproteins. A step of determining multiple peptide-HLA immunogenicity metrics for each peptide sequence in the first peptide set, A step of determining whether each peptide sequence in the first peptide set has a peptide-HLA immunogenicity metric that satisfies a threshold with respect to at least three HLA alleles. A step of producing a second peptide set comprising the first peptide set and a plurality of modified peptide sequences, wherein each of the plurality of modified peptide sequences comprises the substitution of at least one amino acid residue of a peptide sequence in the first peptide set. A step of determining multiple peptide-HLA immunogenicity metrics for each peptide sequence in the second peptide set, and A step of creating a third peptide set by selecting a subset of the second peptide set, wherein the selection includes calculating predicted vaccine performance, the calculation of predicted vaccine performance includes excluding the peptide-HLA immunogenicity metric of the modified peptide sequence with respect to the first HLA allele if the peptide-HLA immunogenicity metric of the unmodified peptide sequence associated with the modified peptide sequence does not meet the threshold with respect to the first HLA allele, and the predicted vaccine performance is a function of the peptide-HLA immunogenicity metric of each peptide sequence in the third peptide set. To implement; Perform experimental assays to obtain a peptide-HLA immunogenicity metric for at least one peptide sequence of the third peptide set; and Forming an immunogenic peptide composition comprising at least one peptide sequence of the third peptide set on which the experimental assay was performed. A method that includes this. [Section 106] The method according to item 105, wherein selecting the plurality of unmodified peptide sequences to produce the first peptide set comprises sliding a window of size n over at least a portion of the amino acid sequences encoding the tumor neoantigen, pathogen proteome, or self-protein, where n is about 8 amino acid lengths to about 25 amino acid lengths, and n is the length of each peptide sequence of the plurality of unmodified peptide sequences in the first peptide set. [Section 107] The method according to item 105, wherein each peptide sequence of the first peptide set is bound to an HLA class I molecule or an HLA class II molecule. [Section 108] The method according to item 105, further comprising filtering the first set of peptides to exclude peptide sequences having a predictive binding core containing a target amino acid residue at an anchor position. [Section 109] The method according to item 105, further comprising substituting at least one amino acid residue in each peptide sequence of the first peptide set. [Section 110] The method according to item 105, wherein the threshold is a binding affinity of less than approximately 1000 nM. [Section 111] The method described in item 105 above, wherein the aforementioned at least three HLA alleles are present in the target HLA type. [Section 112] The method according to item 111, wherein the plurality of peptide sequences are derived from the tumor neoantigen, the pathogen proteome, or the autoprotein present in the subject. [Section 113] The method according to item 105, wherein the immunogenic peptide composition comprises a nucleic acid sequence encoding the amino acid sequence of at least one peptide sequence of the third peptide set. [Section 114] A method for forming an immunogenic peptide composition, Using the processor, A step of creating a first peptide set by selecting multiple unmodified peptide sequences, wherein each peptide sequence of the multiple unmodified peptide sequences is associated with a tumor neoantigen, a pathogen proteome, or an autoprotein. A step of determining multiple peptide-HLA binding scores for each peptide sequence in the first peptide set, A step of determining whether each peptide sequence in the first peptide set has a peptide-HLA binding score that satisfies a threshold with respect to at least three HLA alleles. A step of producing a second peptide set comprising the first peptide set and a plurality of modified peptide sequences, wherein each of the plurality of modified peptide sequences comprises the substitution of at least one amino acid residue of a peptide sequence in the first peptide set. A step of determining multiple peptide-HLA binding scores for each peptide sequence in the second peptide set, and A step of creating a third peptide set by selecting a subset of the second peptide set, wherein the selection includes calculating predicted vaccine performance, the calculation of predicted vaccine performance includes excluding the peptide-HLA binding score of the modified peptide sequence with respect to the first HLA allele if the peptide-HLA binding score of the unmodified peptide sequence associated with the modified peptide sequence does not meet the threshold with respect to the first HLA allele, and the predicted vaccine performance is a function of the peptide-HLA binding scores of each peptide sequence in the third peptide set. To implement; Perform experimental assays to obtain a peptide-HLA immunogenicity metric for at least one peptide sequence of the third peptide set; and Forming an immunogenic peptide composition comprising at least one peptide sequence of the third peptide set on which the experimental assay was performed. A method that includes this. [Section 115] A second threshold is determined by the method described in item 114 above, based on data obtained from one or more experimental assays. [Section 116] The method according to item 114, wherein the predicted vaccine performance is further a function of the peptide-HLA immunogenicity metric of at least one modified peptide sequence of the second peptide set with respect to the second HLA allele, where the first peptide sequence of the first peptide set is predicted to bind to the second HLA allele by a first binding core, the first binding core is the binding core of the first peptide sequence, the first binding core is identical to the second binding core, the first binding core and the second binding core each contain an amino acid position within the peptide sequence, and the second binding core is the binding core of the at least one modified peptide sequence. [Section 117] The method according to item 114, wherein the plurality of peptide sequences are derived from the tumor neoantigen, the pathogen proteome, or the autoprotein present in the target. [Section 118] A method for forming an immunogenic peptide composition, Using the processor, A step of creating a first peptide set by selecting multiple unmodified peptide sequences, wherein at least one of the multiple unmodified peptide sequences is associated with a tumor neoantigen, a pathogen proteome, or an autoprotein. A step of determining multiple peptide-HLA binding scores for each peptide sequence in the first peptide set, A step of determining whether each peptide sequence in the first peptide set has a peptide-HLA binding score that satisfies a threshold with respect to at least three HLA alleles. A step of producing a second peptide set comprising the first peptide set and a plurality of modified peptide sequences, wherein each of the plurality of modified peptide sequences comprises the substitution of at least one amino acid residue of a peptide sequence in the first peptide set. A step of determining multiple peptide-HLA binding scores for each peptide sequence in the second peptide set, and A step of creating a third peptide set by selecting a subset of the second peptide set, wherein the selection includes calculating predicted vaccine performance based on the target HLA type, and the calculation of the predicted vaccine performance includes excluding the peptide-HLA binding score of the modified peptide sequence with respect to the first HLA allele if the peptide-HLA binding score of the unmodified peptide sequence associated with the modified peptide sequence does not meet the threshold with respect to the first HLA allele. To implement; Perform experimental assays to obtain a peptide-HLA immunogenicity metric for at least one peptide sequence of the third peptide set; and Forming an immunogenic peptide composition comprising at least one peptide sequence of the third peptide set on which the experimental assay was performed. A method that includes this. [Section 119] The method according to item 118, wherein each peptide sequence in the first peptide set is bound to an HLA class I molecule or an HLA class II molecule. [Section 120] The method according to item 118, wherein the plurality of peptide sequences are derived from the tumor neoantigen, the pathogen proteome, or the autoprotein present in the subject. The compositions, systems, and methods disclosed herein are not limited to the scope of the specific embodiments described herein. In fact, those skilled in the art will see from the above description that various modifications of the compositions, systems, and methods are not limited to those described herein.

Claims

1. A composition comprising one or more polynucleotides encoding fragments of the KRAS protein, wherein the one or more polynucleotides encode at least two amino acid sequences selected from the group consisting of SEQ ID NOs: 1, SEQ ID NOs: 2, SEQ ID NOs: 21, SEQ ID NOs: 23, SEQ ID NOs: 86, SEQ ID NOs: 88, SEQ ID NOs: 125, SEQ ID NOs: 154, SEQ ID NOs: 194, SEQ ID NOs: 200, and SEQ ID NOs:

366.

2. The composition according to claim 1, wherein the one or more polynucleotides are included in a construct for in vivo expression of at least two peptides encoded by the one or more polynucleotides in a subject.

3. The composition according to claim 2, wherein the administration of one or more polynucleotides in the subject results in the presentation of each of the at least two peptides encoded by the one or more polynucleotides by one or more HLA class I molecules.

4. The administration of one or more polynucleotides is Presentation of the first peptide of the at least two peptides by a first plurality of HLA class I alleles, and Presentation of the second peptide of the at least two peptides by a second plurality of HLA class I alleles This results in the first plurality of HLA class I alleles and the second plurality of HLA class I alleles having at least one different HLA class I allele. The composition according to claim 3.

5. A composition according to any one of claims 1 to 4, which is immunogenic and for the prevention or treatment of cancer in a subject.

6. The composition according to claim 3, wherein each of the at least two peptides is presented by the one or more HLA class I molecules with a peptide-HLA binding affinity value of less than about 50 nM.

7. A composition comprising one or more polynucleotides encoding a fragment of the KRAS protein, wherein the one or more polynucleotides encode at least two amino acid sequences selected from the group consisting of SEQ ID NOs: 3, SEQ ID NOs: 4, SEQ ID NOs: 6, SEQ ID NOs: 15, SEQ ID NOs: 16, SEQ ID NOs: 69, SEQ ID NOs: 72, SEQ ID NOs: 118, SEQ ID NOs: 122, SEQ ID NOs: 131, SEQ ID NOs: 138, SEQ ID NOs: 152, SEQ ID NOs: 189, SEQ ID NOs: 233, SEQ ID NOs: 353, SEQ ID NOs: 514, and SEQ ID NOs:

591.

8. The composition according to claim 7, wherein the one or more polynucleotides are included in a construct for in vivo expression of at least two peptides encoded by the one or more polynucleotides in a subject.

9. The composition according to claim 8, wherein the administration of one or more polynucleotides in the subject results in the presentation of each of the at least two peptides encoded by the one or more polynucleotides by one or more HLA class I molecules.

10. The administration of one or more polynucleotides is Presentation of the first peptide of the at least two peptides by a first plurality of HLA class I alleles, and Presentation of the second peptide of the at least two peptides by a second plurality of HLA class I alleles This results in the first plurality of HLA class I alleles and the second plurality of HLA class I alleles having at least one different HLA class I allele. The composition according to claim 9.

11. A composition according to any one of claims 7 to 10, which is immunogenic and for the prevention or treatment of cancer in a subject.

12. The composition according to claim 9, wherein each of the at least two peptides is presented by the one or more HLA class I molecules with a peptide-HLA binding affinity value of less than about 50 nM.

13. A composition comprising one or more polynucleotides encoding fragments of the KRAS protein, wherein the one or more polynucleotides encode at least two amino acid sequences selected from the group consisting of SEQ ID NOs: 8, SEQ ID NOs: 9, SEQ ID NOs: 10, SEQ ID NOs: 17, SEQ ID NOs: 18, SEQ ID NOs: 20, SEQ ID NOs: 25, SEQ ID NOs: 113, SEQ ID NOs: 181, SEQ ID NOs: 195, SEQ ID NOs: 212, and SEQ ID NOs:

324.

14. The composition according to claim 13, wherein the one or more polynucleotides are included in a construct for in vivo expression of at least two peptides encoded by the one or more polynucleotides in a subject.

15. The composition according to claim 14, wherein the administration of one or more polynucleotides in the subject results in the presentation of each of the at least two peptides encoded by the one or more polynucleotides by one or more HLA class I molecules.

16. The administration of one or more polynucleotides is Presentation of the first peptide of the at least two peptides by a first plurality of HLA class I alleles, and Presentation of the second peptide of the at least two peptides by a second plurality of HLA class I alleles This results in the first plurality of HLA class I alleles and the second plurality of HLA class I alleles having at least one different HLA class I allele. The composition according to claim 15.

17. A composition according to any one of claims 13 to 16, which is immunogenic and for the prevention or treatment of cancer in a subject.

18. The composition according to claim 15, wherein each of the at least two peptides is presented by the one or more HLA class I molecules with a peptide-HLA binding affinity value of less than about 50 nM.

19. A composition comprising one or more polynucleotides encoding a fragment of the KRAS protein, wherein the one or more polynucleotides encode at least two amino acid sequences selected from the group consisting of SEQ ID NOs: 26, 27, 28, 30, 36, 37, 70, 71, 73, 74, 78, 176, and 187.

20. The composition according to claim 19, wherein the one or more polynucleotides are included in a construct for in vivo expression of at least two peptides encoded by the one or more polynucleotides in a subject.

21. The composition according to claim 20, wherein the administration of one or more polynucleotides in the subject results in the presentation of each of the at least two peptides encoded by the one or more polynucleotides by one or more HLA class I molecules.

22. The administration of one or more polynucleotides is Presentation of the first peptide of the at least two peptides by a first plurality of HLA class I alleles, and Presentation of the second peptide of the at least two peptides by a second plurality of HLA class I alleles This results in the first plurality of HLA class I alleles and the second plurality of HLA class I alleles having at least one different HLA class I allele. The composition according to claim 21.

23. A composition according to any one of claims 19 to 22, which is immunogenic and for the prevention or treatment of cancer in a subject.

24. The composition according to claim 21, wherein each of the at least two peptides is presented by the one or more HLA class I molecules with a peptide-HLA binding affinity value of less than about 50 nM.

Images