Bioinformatic processes for determination of peptide binding
This invention relates to the identification of peptide binding to ligands, and in particular to identification of epitopes expressed by microorganisms and by mammalian cells. The present invention provides polypeptides comprising the epitopes, and vaccines, antibodies and diagnostic products that utilize or are developed using the epitopes.
1. A process of identifying variant cancer cell epitopes that bind to an MHC binding region and producing a vaccine comprising:
a) obtaining amino acid sequences for a cancer cell protein;
b) providing peptide binding prediction equations for multiple MHC binding regions derived by:
(i) assembling experimentally derived data comprising a multiplicity of measurements of amino acid physicochemical properties;
(ii) producing a correlation matrix of the experimentally derived data;
(iii) deriving by Principal Component Analysis multiple uncorrelated dimensionless, weighted and ranked proxy descriptors to describe at least 80% of the variance in said physicochemical properties of individual amino acids,
(iv) using said proxy descriptors to describe individual amino acids in peptides with known binding affinities thereby creating vectors which describe said peptides with known binding affinities, and
(v) via neural network modeling, deriving said binding prediction equations by comparing said vectors with said known binding affinities of said peptides;
(c) in-putting said amino acid sequences from said cancer cell protein into a computer;
(d) identifying peptides of interest in said cancer cell protein, wherein the peptides of interest are suspected cancer cell epitopes, and determining the binding affinity of the peptides of interest by applying the MHC binding prediction equations;
(e) generating in silico at least 1000 variants of the peptides of interest by substituting individual amino acids of the peptides with a different amino acid;
(f) applying proxy descriptors derived from the Principal Component Analysis of step (b) to describe each of the variant peptides generated in step (e) and applying the MHC binding prediction equations of step (b) to each of the variant peptides to determine the predicted MHC binding of each variant peptide;
(g) comparing the predicted binding to an MHC binding region of each of the variant peptides generated in step (e) with the binding affinity of the peptides of interest in the cancer protein to identify variant peptide sequences with higher or lower predicted binding affinity to said MHC binding region;
(h) synthesizing one or more of said variant peptide sequences which are identified as having a higher or lower predicted binding affinity to said MHC binding region;
(i) formulating the synthesized peptides with a pharmaceutical carrier to provide a vaccine, and
(j) administering the vaccine to a subject that has cancer under conditions such that an immune response is generated.
2. The process of claim 1 wherein said synthesized peptide is identified as having higher or lower binding affinity to more than one binding partner.
3. The process of claim 1 , wherein said binding affinity of said variant peptide sequence is increased as compared to the binding affinity of the peptide of interest in the cancer protein.
4. The process of claim 1 , wherein said binding affinity of said variant peptide sequence is decreased as compared to the binding affinity of the peptide of interest in the cancer protein.
5. The process of claim 1 , wherein said binding affinity of the variant peptide sequence is increased or decreased by >1 standard deviation units.
6. The process of claim 1 , wherein said binding affinity of the variant peptide sequence is increased or decreased by >2 standard deviation units.
7. The process of claim 1 , wherein from about 10 to about 100 peptide sequences are selected for synthesis.
8. The process of claim 1 , wherein at least 2 amino acids of said synthesized peptide sequence are modified.
9. The process of claim 1 , wherein from 1,000 to about 5,000 variant peptide sequences are generated in step (e).
10. The process of claim 1 , wherein from about 5,000 to about 10,000 variant peptide sequences are generated in step (e).
11. The process of claim 1 , wherein from about 1 to about 10 peptide sequences are selected for synthesis in step (h).