Computer assisted antibody re-epitoping

1. A method for generating a library of variant antigen binding molecules for screening for binding to an epitope of interest, said method comprising:

(a) analyzing binding affinities of a set of possible template antigen binding molecules for the epitope of interest and selecting sequence data of a template antigen-binding molecule from said set of possible template antigen binding molecules, wherein said selected template does not bind the epitope of interest and wherein said set of possible template antigen-binding molecules consists of a plurality of known antibodies that do not bind the epitope of interest, wherein said selecting sequence data comprises screening three-dimensional structures or structural models of the set of possible template antigen-binding molecules based on one or more of the following criteria: shape complementarity to the epitope of interest, physico-chemical complementarity to the epitope of interest and the predicted free energy of the interaction with the epitope of interest;

(b) selecting at least one residue position in said selected sequence data of the template antigen-binding molecule of (a) for mutation, wherein said selecting said at least one position comprises (i) screening the three dimensional structure and/or a three-dimensional model of the template antigen-binding molecule sequence data selected in (a) to identify residues that contribute to binding to the epitope of interest; or (ii) conducting multiple sequence alignments of the nucleic acid or the amino acid sequence of the template antigen-binding molecule selected in (a) to identify substitutable positions;

(c) selecting at least one variant residue to substitute at the at least one residue position selected in the sequence date selected in (b), wherein said selecting the at least one variant residue comprises for each residue identified in step (b), identifying one or more amino acid substitutions that are preferred, allowed or neutral at that residue position;

(d) substituting at least one variant residue at the at least one residue position selected in (c) in said sequence data;

(e) generating sequence data of one or more variant antigen binding molecules and synthesizing one or more variant antigen binding molecules of said template antigen-binding molecule, wherein said sequence data of the variant antigen binding molecules comprises at least one substitution identified and substituted in the sequence data of (d); and

(f) screening said variant antigen binding molecules synthesized, said molecules comprising antibodies or antigen-binding fragments thereof, for binding to the epitope of interest, and selecting variant antigen binding molecules that bind the epitope of interest;

wherein said method generates a library of variant antigen-binding molecules comprising said substituted sequence data of said template antigen-binding molecule for screening for binding to the epitope of interest.

2. The method of claim 1 , wherein said preferred, allowed and/or neutral substitutions in the sequence data are determined by analyzing the amino acid sequences of a plurality of known antibodies compared with the sequence data of the template antigen-binding molecule, wherein analyzing comprises comparison of said amino acid sequences, analysis of composition of said amino acid sequences, analysis of ΔΔG of binding energy, probability of mutation from said germline sequence, or sequence-similarity search algorithms, or any combination thereof.

3. The method of claim 2 , wherein said substitutions in the sequence data comprise

a position prone to somatic hypermutation;

a position selected based on sequence analysis using multi-sequence alignment to the template, wherein said analysis is by machine learning; or

a position based on structural analysis, wherein said analysis is by machine learning; or

a combination thereof.

4. The method of claim 1 , wherein said at least one residue selected for mutation is in a CDR region of the sequence data of the template.

5. The method of claim 1 , wherein the at least one variant residue selected for mutation comprises residues that is in less than all of the CDRs.

6. The method of claim 1 wherein said method is computer implemented.