ABSTRACT
In the structural biology of proteins, the dogma is that sequence denes structure, which in turn denes function. In order to design functional proteins, therefore, the rst step is to ll in the gap between sequence and structure; if a sequence can be designed for an arbitrary structure, then function should follow. To this end, many design algorithms have focused on the ability to reliably predict sequences that can fold, specically into a desired tertiary structure. This has proven to be a robust test of our understanding of the forces that govern protein folding.
