Protein refinement

At atomic resolution, the number of observables is much higher than at medium or low resolution (see Table 10.1) allowing for the refinement of models with more parameters representing a more detailed description of the structure and the flexibility of the molecule in the crystal, for example by using anisotropic displacements parameters (17, U, 17, instead of isotropic 5-values that are 

Different aspects of atomic resolution structures of biological macromolecules have been reviewed (e.g. Schmidt and Lamzin, 2002 Esposito et ai, 2002 Vrielink and Sampson, 2003). Examples of particular interest are  

In the following, the typical steps of the refinement of a protein structure at atomic resolution using SHELXL will be described. The functionality discussed can be 

When macromolecular stmctures are refined at atomic resolution, a number of concepts that are commonly used in the refinement of small molecules have to be introduced. These include parameterizations for modeling the static and dynamic disorder of atoms in a crystal and for the positioning of hydrogens. Furthermore, the treatment of both ordered and bulk solvent and the determination of standard uncertainties via the inversion of the normal matrix of the refinement need to be discussed. 

The introduction of aiusotropic displacements parameters (or ADPs) into a refinement more than doubles the number of parameters from 4 (3 coordinates plus 1 5-value) to 9 (3 coordinates plus 6 ADPs) per atom and thus needs to be tightly monitored to avoid over-fitting the data. Typically, a drop in 5free of at least 1.0-1.5% should be observed upon switching from isotropic to anisotropic displacement parameters. In many cases, the inclusion of ADPs results in a dramatic improvement of the crystallographic phases and the corresponding electron density maps. 

---

Sack, J. S., Trakhanov, S. D., Tsigannik, I. H., and Quiocho, F. A. (1989b). Structure of the L-leucine-binding protein refined at 2.4 A resolution and comparison with the Leu/lle/Val-binding protein structure. J. Mol. Biol. 206, 193-207. 

This chapter is structured as follows In Sect. 6.2, a basic introduction to molecular refinement is presented, stressing particularly relevant aspects. Section 6.3 reviews the recent work by Falklof et al., describing how the 2 x 2 x 2 supercell for the lysozyme structure was obtained. Section 6.4 reviews some modern advances in DFT, focusing on dispersion-corrected DFT, while Sect. 6.5 describes the effects of DFT optimization of atomic coordinates on the agreement between observed and calculated X-ray structure factors. The aim is to determine an optimal electronic-structure computational procedure for quantum protein refinement, and we consider only the effects of minor local perturbations to the existing protein model rather than those that would be produced by allowing full protein refinement. 

One third of the GMTKN30 database consists of test sets dealing with non-covalent interactions (e.g., intermolecular and intramolecular London dispersion, hydrogen bonds), interactions that are of utmost importance in a protein refinement. Table 6.1 shows the best functionals for non-covalent interactions for rungs two to four of the five current rungs of Perdew s ladder scheme. 

It has already been mentioned that the PW91 and BP86 methods will be tested owing to their widespread popularity. These methods are often applied using Pople s 6-31G basis set [111] and so we also examine these methods/basis-set combinations for their usefulness in protein refinement. Parameters for BP86-D3(BJ)/6-31G are also shown in Table 6.2. Table 6.3 gives an overview of the various levels of theory tested in this study. 

Because of the limited resolution of X-ray data for proteins, refinements that take some account of anisotropic motions have introduced assumptions concerning the nature of the anisotropy. One possibility is to assume anisotropic rigid body motions for sidechains such as tryphophan and phenyl-... 

Most chapters were written by me. The chapter on twinning is by Regine Herbst-Irmer, that on protein refinement by Thomas Schneider, the chapter about small-molecule structure validation by Ton Speck, and that on protein structure validation by Michael Sawaya. George Sheldrick, author of SHELXL, wrote the foreword to this book, which includes a brief history of SHELXL. 

Cowan, S. W., Newcomer, M, E., and Jones, T. A. (1993) Crystallographic studies on a family of cellular lipophilic transport proteins-refinement of P2 myelin... 

---