Molecular Replacement for Multi-Domain Structures Using Packing Models

Abstract

Molecular replacement (MR) is frequently used to obtain phase information for a unit cell packed with a macromolecule of unknown structure. The goal of MR searches is to place a homologous/similar molecule in the unit cell so as to maximize the correlation with x-ray diffraction data. MR software packages typically perform rotation and translation searches separately. This works quite well for single-domain proteins. However, for multi-domain structures and complexes, computational requirements can become prohibitive and the desired peaks can become hidden in a noisy landscape.The main contribution of our approach is that computationally expensive MR searches in continuous configuration space are replaced by a search on a relatively small discrete set of representative packing arrangements of a multi-rigid-body model. These arrangements are generated by minimizing a potential function that forces the model conformations to separate from each other and not overlap within the unit cell. This is done before computing Patterson correlations rather than only performing collision checks when evaluating the feasibility of peaks. The list of feasible arrangements is short because packing constraints together with unit cell symmetry and geometry impose strong constraints.In numerical trials, we found that a candidate from the feasible set is usually similar to the arrangement of the target structure within the unit cell. To further improve the accuracy, a refined search can be performed in the neighborhood of this packing arrangement. Our approach is demonstrated with multi-domain models in silico for both 2D and 3D, with ellipsoids representing both the domains of the model and target structures. Our results show that an approximate phase can be found with 5 percent error and significantly improved search speed.We acknowledge NIH Grant R01GM075310 for the support of this work and Dr. E. Lattman for useful discussions.