[Molvis-list] exporting _good_ h-bonds (was: export Chime h-bonds)

Gale Rhodes rhodes at usm.maine.edu
Tue Jan 3 11:20:28 EST 2006

Hi, Tim!

On Jan 3, 2006, at 10:57 AM, Timothy Driscoll wrote:

> hi Kevin,
> thanks for the clarification.  how would you define a 'high- 
> quality' x-ray structure?  is there a generally accepted threshold  
> for resolution or RMSD?

Resolution and RMSD on bonds and angles tell you very little about  
the quality of models. For example, a 3.0 A structure with non- 
crystallographic averaging might be of higher quality than a 2.2 A  
structure without it. Any parameters, like bond lengths and angles,  
that are constrained or restrained during refinement are poor  
indicators of quality.

To become as astute judge of model quality, take Gerard Kleywegt's  
tutorial on the subject:


I know of no better or more state-of-the-art introduction to the  
subject. This guide helps model users understand that the pertinent  
quality criteria depend on what you want to do with the model.

Chapter 8 of my book, Crystallography Made Crystal Clear, is another  
source of simple explanations, but Kleywegt puts you in the driver's  
seat of some of the best vehicles for assessing quality.

Hope this helps.


Gale Rhodes, Professor of Chemistry
University of Southern Maine
PO Box 9300
Portland, Maine 04104-9300

More information about the Molvis-list mailing list