The stability of model proteins with designed sequences is assessed in terms of the number of sequences, obtained from the designed sequence through mutations, which fold into the “native” conformation. By a complete enumeration of the sequences obtained by introducing up to four point mutations and up to seven composition-conserving mutations (swapping of amino acids) in a 36mers chain and by running dynamic simulations on the mutated sequences, it is found that this number depends on the gap between the native conformation and the bulk of misfolded conformations, but not on the particular designed sequence, provided its associated energy gap is large.

• Received 28 September 1998

DOI:https://doi.org/10.1103/PhysRevLett.82.4727