Determination of protein structures--a series of fortunate events

Abstract

Determination of a macromolecular structure using x-ray diffraction is a multistep process that involves a plethora of techniques involving molecular biology, bioinformatics, and physical sciences. Counterintuitively, the success of any or all individual steps does not guarantee the success of the overall process. This review examines the difficulties presented by each step on the path from a gene to the final publication, together with certain lucky (or unlucky) circumstances that can affect the velocity along that path.

FIGURE 1

The fraction of x-ray structures deposited in the PDB that report the use of synchrotron sources for their determination (dark blue) and sample temperatures close to 100 K during data collection (light blue).

FIGURE 2

Residue 160 in the crystal structure of soybean lipoxygenase shown together with 2F_o-F_c electron density map. (A) A map contoured at 1σ level. The crystal was obtained using natural protein isolated from soybeans and the residue was identified as serine (PDB codes: 1YGE and 1F8N). (B) Map shown at 0.7σ. The serine was replaced by glutamic acid, what is consistent with results of the DNA sequencing (Ted Holman, 2007, private communication). (C) Electron density shown at 0.3σ. Different contouring of the map reveals either possible decarboxylation during data collection or conformational flexibility of Glu¹⁶⁰.

FIGURE 3

Comparison of the extent of application of SAD and MAD techniques in APS and ESRF, as reported in PDB.