Designing new peptide and protein therapeutics using adiabatic quantum annealers*

The ability to design new peptide and protein sequences that can adopt folds not found in nature permits new functions to be engineered.  Of particular interest is the rational design of peptide and protein therapeutics that are able to bind specifically to, and alter the function of, target biomolecules implicated in human disease.  Over the last 20 years, major advancements have been made in classical computing approaches for peptide and protein design.  Nevertheless, the design problem is NP-complete: no known classical algorithm scales well as either the number of amino acids in the polypeptide chain grows large (as in the case of large proteins) or the number of building-blocks from which one may choose grows large (as in the case of synthetic peptides that can be built from hundreds of artificial amino acid types).  This limits the complexity of the design problems that can currently be solved classically.  Here, we introduce approaches for designing peptides and proteins using adiabatic quantum annealers to solve the hard combinatorial problem of design, without simplification or reduction to a toy problem.  Using the D-Wave Advantage system, we demonstrate application of our methods to nontrivial peptide design problems.