Method and systems for prediction of hla class ii-specific epitopes and characterization of cd4+ t cells

Abstract

Methods for preparing a personalized cancer vaccine and a method to train a machine-learning HLA-peptide presentation prediction model.

1 .- 61 . (canceled) 
     
     
         62 . A system for selecting one or more peptide sequences for preparing a pharmaceutical composition, the system comprising a computer processor comprising:
 (a) an input module configured to receive amino acid sequence information of a set of candidate peptide sequences expressed by cells of a human subject, wherein each candidate peptide sequence of the plurality of candidate peptide sequences is encoded by a genome, transcriptome, or exome of a human subject, or a pathogen or a virus in the human subject;   (b) a processing module operably linked to the input module, the processing module comprising an executable code comprising a trained machine learning class II HLA-peptide presentation prediction model, wherein the trained machine learning class II HLA-peptide presentation prediction model is configured to generate output peptide sequences with a plurality of presentation predictions, wherein each presentation prediction of the plurality of presentation predictions is indicative of a presentation likelihood that a peptide sequence of the set of candidate peptide sequences is presented by one or more proteins encoded by a class II HLA allele of a cell of the human subject, and wherein the trained machine learning class II HLA-peptide presentation prediction model comprises;
 (i) a plurality of parameters identified at least based on training data comprising:
 (1) sequences of training peptides, 
 (2) an identity of a protein encoded by an HLA class II allele associated with the training peptide sequences, and 
 (3) an observation by mass spectrometry that one or more of the training peptides was presented by the protein encoded by the HLA class II allele in training cells; and 
 
 (ii) a function representing a relation between the amino acid sequence information received as input and the presentation likelihood generated as an output based on the amino acid sequence information and the plurality of parameters; 
   
       wherein each peptide sequence of a subset of the output peptide sequences is for preparing a therapeutic composition for the human subject based on the presentation likelihood generated as the output of the peptide sequence being in a complex with the one or more proteins encoded by a class II HLA allele of a cell of the human subject, and 
       wherein the trained machine learning class II HLA-peptide presentation prediction model has a positive predictive value (PPV) of at least 0.2 according to a presentation PPV determination method. 
     
     
         63 . The system of  claim 62 , wherein the input module is configured to receive an identity of one or more proteins encoded by a class II HLA allele of a cell of the human subject. 
     
     
         64 . The system of  claim 62 , wherein the processing module is linked to an output module configured to display the plurality of presentation predictions. 
     
     
         65 . The system of  claim 62 , wherein the trained machine learning class II HLA-peptide presentation prediction model has a PPV of at least 0.2 when amino acid information of a plurality of test peptide sequences are processed to generate a plurality of test presentation predictions, each test presentation prediction indicative of a likelihood that the one or more proteins encoded by a class II HLA allele of a cell of the subject can present a given test peptide sequence of the plurality of test peptide sequences, wherein the plurality of test peptide sequences comprises at least 500 test peptide sequences comprising:
 (i) at least one hit peptide sequence identified by mass spectrometry to be presented by an HLA protein expressed in cells, and   (ii) at least 499 decoy peptide sequences contained within a protein encoded by a genome of an organism, wherein the organism and the subject are the same species,   
       wherein the plurality of test peptide sequences comprises a ratio of 1:499 of the at least one hit peptide sequence to the at least 499 decoy peptide sequences and a top 0.2% of the plurality of test peptide sequences are predicted to be presented by the HLA protein expressed in cells by the trained machine learning class II HLA-peptide presentation prediction model. 
     
     
         66 . The system of  claim 62 , wherein:
 (i) the at least one hit peptide sequence comprises at least 10 hit peptide sequences, and   (ii) the at least 499 decoy peptide sequences comprise at least 4990 decoy peptide sequences.   
     
     
         67 . The system of  claim 62 , wherein any nine contiguous amino acid subsequences of any of the at least one hit peptides does not overlap with any nine contiguous amino acid subsequences of the at least 4990 decoy peptide sequences. 
     
     
         68 . The system of  claim 62 , wherein the trained machine learning class II HLA-peptide presentation prediction model has a PPV of at least 0.2 at a recall rate of 10% according to a presentation PPV determination method. 
     
     
         69 . The system of  claim 62 , wherein the trained machine learning class II HLA-peptide presentation prediction model has a PPV of at least 0.2 at a recall rate of 20% according to a presentation PPV determination method. 
     
     
         70 . The system of  claim 62 , wherein the trained machine learning class II HLA-peptide presentation prediction model has a PPV of at least 0.3 according to a presentation PPV determination method. 
     
     
         71 . The system of  claim 62 , wherein the trained machine learning class II HLA-peptide presentation prediction model has a PPV of at least 0.3 at a recall rate of 20% according to a presentation PPV determination method. 
     
     
         72 . The system of  claim 62 , wherein each peptide sequence of the set of candidate peptide sequences is associated with a cancer. 
     
     
         73 . The system of  claim 72 , wherein each peptide sequence of the set of candidate peptide sequences
 (i) comprises a mutation,   (ii) is expressed in a cancer cell of the subject, and   (iii) is not encoded by a genome of a non-cancer cell of the human subject.   
     
     
         74 . The system of  claim 62 , wherein each sequence of the one or more of the training peptides sequences observed by mass spectrometry to be presented by the protein encoded by the HLA class II in training cells has a length of at least 15 amino acids. 
     
     
         75 . The system of  claim 62 , wherein the training cells comprise training cells expressing a single MHC class II complex or a protein encoded by a single allelic variant of a class II HLA locus selected from the group consisting of DR, DP, and DQ, wherein the single MHC class II complex or a protein encoded by the single allelic variant of a class II HLA locus is expressed by a cell of the subject. 
     
     
         76 . The system of  claim 62 , wherein the training data comprises training data obtained by deconvolution. 
     
     
         77 . The system of  claim 62 , wherein the training cells express a protein encoded by a class II HLA allele of a cell of the human subject, wherein the protein encoded by a class II HLA allele comprises an affinity tag. 
     
     
         78 . The system of  claim 62 , wherein the protein encoded by a class II HLA allele is selected from the group consisting of: HLA-DPB1*01:01/HLA-DPA1*01:03, HLA-DPB1*02:01/HLA-DPA1*01:03, HLA-DPB1*03:01/HLA-DPA1*01:03, HLA-DPB1*04:01/HLA-DPA1*01:03, HLA-DPB1*04:02/HLA-DPA1*01:03, HLA-DPB1*06:01/HLA-DPA1*01:03, HLA-DRB1*01:01, HLA-DRB1*01:02, HLA-DRB1*03:01, HLA-DRB1*03:02, HLA-DRB1*04:01, HLA-DRB1*04:02, HLA-DRB1*04:03, HLA-DRB1*04:04, HLA-DRB1*04:05, HLA-DRB1*04:07, HLA-DRB1*07:01, HLA-DRB1*08:01, HLA-DRB1*08:02, HLA-DRB1*08:03, HLA-DRB1*08:04, HLA-DRB1*09:01, HLA-DRB1*10:01, HLA-DRB1*11:01, HLA-DRB1*11:02, HLA-DRB1*11:04, HLA-DRB1*12:01, HLA-DRB1*12:02, HLA-DRB1*13:01, HLA-DRB1*13:02, HLA-DRB1*13:03, HLA-DRB1*14:01, HLA-DRB1*15:01, HLA-DRB1*15:02, HLA-DRB1*15:03, HLA-DRB1*16:01, HLA-DRB3*01:01, HLA-DRB3*02:02, HLA-DRB3*03:01, HLA-DRB4*01:01, HLA-DRB5*01:01, HLA-DRB1*01:01, HLA-DRB1*01:02, HLA-DRB1*03:01, HLA-DRB1*04:01, HLA-DRB1*04:02, HLA-DRB1*04:04, HLA-DRB1*04:05, HLA-DRB1*07:01, HLA-DRB1*08:01, HLA-DRB1*08:02, HLA-DRB1*08:03, HLA-DRB1*09:01, HLA-DRB1*11:01, HLA-DRB1*11:02, HLA-DRB1*11:04, HLA-DRB1*12:01, HLA-DRB1*13:01, HLA-DRB1*13:02, HLA-DRB1*13:03, HLA-DRB1*14:01, HLA-DRB1*15:01, HLA-DRB1*15:02, HLA-DRB1*15:03, HLA-DRB1*16:02, HLA-DRB3*01:01, HLA-DRB3*02:01, HLA-DRB3*02:02, HLA-DRB3*03:01, HLA-DRB4*01:01, HLA-DRB4*01:03, HLA-DRB5*01:01; HLA-DPB1*01:01, HLA-DPB1*02:01, HLA-DPB1*02:02, HLA-DPB1*03:01, HLA-DPB1*04:01, HLA-DPB1*04:02, HLA-DPB1*05:01, HLA-DPB1*06:01, HLA-DPB1*11:01, HLA-DPB1*13:01, HLA-DPB1*17:01, HLA-DQA1*01:01/HLA-DQB1*05:01, HLA-DQA1*01:02/HLA-DQB1*06:02, HLA-DQA1*01:02/HLA-DQB1*06:04, HLA-DQA1*01:03/HLA-DQB1*06:03, HLA-DQA1*02:01/HLA-DQB1*02:02, HLA-DQA1*02:01/HLA-DQB1*03:03, HLA-DQA1*03:01/HLA-DQB1*03:02, HLA-DQA1*03:03/HLA-DQB1*03:01, HLA-DQA1*05:01/HLA-DQB1*02:01, HLA-DQA1*05:05/HLA-DQB1*03:01, and any combination thereof. 
     
     
         79 . The system of  claim 62 , wherein each peptide sequence of the subset of the output peptide sequences binds to a protein encoded by a class II HLA allele of a cell of the human subject with an IC50 of 500 nM or less, or a predicted IC50 of 500 nM or less. 
     
     
         80 . The system of  claim 62 , wherein each peptide sequence of the subset of the output peptide sequences is for preparing a therapeutic composition for the human subject that comprises one or more polypeptides comprising at least two peptide sequence of the subset of the output peptide sequences or one or more polynucleotides encoding at least two of peptide sequence of the subset of the output peptide sequences 
     
     
         81 . The system of  claim 62 , wherein each candidate peptide sequence of the plurality of candidate peptide sequences is encoded by a genome, transcriptome, or exome of a human subject with cancer.