Quantitative fitting of atomic models into observed densities derived by electron microscopy.

A new methodology for fitting atomic models into density distributions is described. This approach is based on a global density correlation analysis that can be optionally supplemented by biochemical as well as biophysical data. The procedure is completely general and enables an objective evaluation of the resulting docking in the light of available biochemical and biophysical information as well as density correlation alone. In this paper we describe the implementation of the algorithm and its application to two biological systems. In both cases the procedure provided an interface model on the atomic level and located parts of the structure that were missing in the atomic model but present in the electron-microscopic construct. It also detected and quantified conformational changes in actomyosin complexes.