The de novo protein design framework is depicted below. Stage One and Stage Two utilize a number of packages available commercially or developed by others within the academic community. To learn more about the packages used, click on the corresponding link below the figure.

Package Description
Naccess 2.1.1 Used to define the mutation set based upon Solvent Accessible Surface Area of each residue.
GAMS 23.3 Used to solve the mixed-integer linear models to global optimiality in Stage One, generating a rank-ordered list of sequences.
CYANA 2.1 Used to generate structures around a defined flexible template in the Stage Two Fold Specificity calculations.
TINKER 3.6 Used to minimize the structures generated using Cyana 2.1 in the Stage Two: Fold Specificity calculations.
Rosetta 2.3 Rosetta 3.3 Used to generate rotamer ensembles of a peptide, protein, and complex in the Stage Two: Approximate Binding Affinity calculations.