PROTEIN-FOLDING - OPTIMIZED SEQUENCES OBTAINED BY SIMULATED BREEDING IN A MINIMALIST MODEL

For a minimalist model of protein folding, which we introduced recentl y, we investigate various methods to obtain folding sequences. A detai led study of random sequences shows that, for this model, such sequenc es usually do not fold to their ground states during simulations. Stra ight-forward techniques for the construction of folding sequences, bas ed solely on the target structure, fail. We describe in detail an opti mization algorithm, based on genetic algorithms, for the ''simulated b reeding'' of folding sequences in this model. We find that, for any ta rget structure studied, there is not only a single folding sequence bu t a patch of sequences in sequence space that fold to this structure. In addition, we show that, much as in veal proteins, nonhomologous seq uences may fold to the same target structure. (C) 1997 John Wiley & So ns, Inc.