A guided simulated annealing method for crystallography.

A new optimization algorithm, the guided simulated annealing method, for use in X-ray crystallographic studies is presented. In the traditional simulated annealing method, the search for the global minimum of a cost function is only determined by the ratio of energy change to the temperature. This method designs a new quality function to guide the search for a minimum. Using a multiresolution process, the method is much more efficient in finding the global minimum than the traditional method. Results for two large molecules, isoleucinomycin (C(60)H(102)N(6)O(18)) and an alkyl calix (C(72)H(112)O(8). 4C(2)H(6)O), with different space groups are reported.