Title:

A Protein Crystallographer's Perspective: Why Our Structures
Aren't Perfect and What We Can Do about It

Author:

Tom Transue
National Institute of Environmental Health Sciences
(NIEHS/NIH)

Abstract:

Macromolecular crystallography has undergone dramatic changes in the last
15 years.  Sample preparation, crystallization, data collection, and
structure determination have all been streamlined and automated for
high-throughput research.  Refinement methods have added cross-validation,
bulk solvent correction, overall anisotropic B-factor adjustment, maximum
likelihood target functions, new graphics program aids, and some automated
scripting, but probably their greatest advance is due to an increase in
computer speeds.  In addition, post-refinement 'validation' has brought to
light a number of features which are not excluded during standard
refinement, but are none-the-less inconsistent with our understanding of
molecular structure.  We seek to identify such features, understand why
they escape detection by current refinement methods, and propose
characteristics of the next generation of refinement methodology. 

oe_cup6_transue.pdf