Resolving Some Old Problems in Protein Crystallography

+ See all authors and affiliations

Science  25 May 2012:
Vol. 336, Issue 6084, pp. 986-987
DOI: 10.1126/science.1222162

You are currently viewing the summary.

View Full Text


Scientific conclusions should be supported by the observed data. However, in x-ray crystallography, the raw diffraction data are rather remotely connected to the final coordinates of the molecule because the experimental data undergo a Fourier transform during the analysis. Thus, any individual feature of the structural model—where a particular atom is located—depends on all of the measured diffraction intensities. Also, the phase information essential for this reconstruction is lost in the experiment (the “phase problem”). Although the coordinate model is repeatedly tested against the data in the course of structure solution, it is common practice to choose what data to use early in the process. Two papers in this issue suggest in different ways that crystallographers have often been excluding useful data from structure determination. On page1030, Karplus and Diederichs (1) show that the “resolution” of data sets is frequently underestimated, so that the final model is not as good as it could be. On page 1033, Liu et al. (2) show that averaging data from multiple crystals can give helpful information for solving the phase problem by using intrinsic sulfur atoms in the protein, circumventing the need to introduce heavier atoms.