Efficient analysis of 51V solid-state MAS NMR spectra using genetic algorithms

A program for iterative fitting procedures to determine the NMR parameters from ⁵¹V solid-state MAS NMR spectra was developed. It contains options to use genetic algorithms and downhill-simplex optimizing procedures to extract the optimal parameter sets, which describe our spectra. As computational kernel the SIMPSON program is employed. Other kernels like SPINEVOLUTION are easily incorporable. The algorithms are checked for their suitability for the present optimization problem and optimal simulation conditions are determined, with the focus on minimal processing time. The procedure leads to a very good agreement between experimental and simulated spectra in a passable period of time. First results for spectra of model compounds for the active site of vanadium haloperoxidases are presented.