The following figure shows what a certain resolution, given in �ngstr�m (�) means for the user of structural models derived from X-ray data. One always has to remember that the cute model you see was built into an experimental electron density. The model may look as good at 3 � as it does at 1.2 �, but is it a correct and unique description of reality ?
The pictures of the electron density at different data set resolution of the same region of a molecule leave no question that a model of a phenylalanine-containing ligand (the 6-ring structure) can be correctly placed into the 1.2 � data. This still can be done with confidence in the 2 � case, but at 3 � we already observe a deviation of the centroid of the ring from the correct model. The bottom panel visualizes the relation between diffraction limit, amount of data and nominal resolution. The more and better data, the more acurate and detailed the final structure model will be.
Most protein crystals diffract between 1.8 and 3 �, a few to very high resolution (the term high resolution is used loosely in macromolecular crystallography, we apply it to data of 1.8-1.2 �, below 'atomic' or 'ultra-high' resolution are commonly used). The most efficient way to increase resolution (short of trying to grow better crystals) is to cryo-cool the crystals to near liquid nitrogen temperature.
You can read the first pages 1, 2, 3 of the introduction of my book Biomolecular Crystallography or buy the book from Amazon. You can also look at the Glossary to find terms.
Back to X-ray Tutorial Index
This World Wide Web site conceived and maintained by Bernhard Rupp. Last revised Dezember 27, 2009 01:40