Abstract
While conventional high-resolution techniques in structural biology are challenged by the size and flexibility of many biological assemblies, recent advances in low-resolution techniques such as cryo-electron microscopy (cryo-EM) and small angle X-ray scattering (SAXS) have opened up new avenues to define the structures of such assemblies. By systematically combining various sources of structural, biochemical and biophysical information, integrative modeling approaches aim to provide a unified structural description of such assemblies, starting from high-resolution structures of the individual components and integrating all available information from low-resolution experimental methods. In this review, we describe integrative modeling approaches, which use complementary data from either cryo-EM or SAXS. Specifically, we focus on the popular molecular dynamics flexible fitting (MDFF) method, which has been widely used for flexible fitting into cryo-EM maps. Second, we describe hybrid molecular dynamics, Rosetta Monte-Carlo and minimum ensemble search (MES) methods that can be used to incorporate SAXS into pseudoatomic structural models. We present concise descriptions of the two methods and their most popular alternatives, along with select illustrative applications to protein/nucleic acid assemblies involved in DNA replication and repair.
Highlights
Molecular Dynamics Flexible FittingThe resolution of current cryo-electron microscopy (EM) methodology is generally not comparable to that of X-ray crystallography [1], cryo-EM is routinely capable of providing coarse structural information on macromolecular complexes, and in a biologically more realistic environment, perhaps even capturing different functional states [32]
Other technologies generate spatial envelopes of biological molecules or assemblies e.g. negative stain electron microscopy (EM) and small angle X-ray scattering (SAXS), while detailed interaction profiles are accessible through methodologies like chemical footprinting, cross-linking, fluorescence resonance energy transfer (FRET), mass spectrometry (MS), proteomics studies, and so on [20,21]
A new xMDFF method for structural determination from low-resolution crystallographic data was introduced, which integrates the functionalities of the original Molecular dynamics flexible fitting (MDFF) method and the PHENIX crystallographic refinement package [51]; The MDFF protocol was modified to work with model-phased densities, wherein experimental X-ray scattering amplitudes are augmented with phases computed from an approximate initial model to produce a density map
Summary
The resolution of current cryo-EM methodology is generally not comparable to that of X-ray crystallography [1], cryo-EM is routinely capable of providing coarse structural information on macromolecular complexes, and in a biologically more realistic environment, perhaps even capturing different functional states [32]. Methods developed for fitting atomic structures into cryo-EM maps can be divided generally into rigid-body docking and flexible fitting. A variety of flexible fitting methods have been developed in recent years, based on different mathematical flavors, including real-space refinement upon segmented rigid-body docking [40,41], normal-mode calculation based on optimization of the correlation between structure and map [42], vector quantization based coarse-grained model fitting [38], etc. Other methods have applied external forces proportional to the gradient of the EM-map (implemented in MDFF) [44] or the gradient of the crosscorrelation coefficient between the structure and the EM-map [45,46] along with MD simulations to guide the atoms into high-density regions of an EM-map. Against high-resolution cryo-EM density maps was developed by the same group, as described below
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
