W0430

Quasi-MAD Phasing. Hongliang Xu, Hauptman-Woodward Medical Research Inst. & Dept. of Structural Biology, SUNY at Buffalo, 73 High St., Buffalo, NY 14203-1196, USA.

MAD (multi-wavelength anomalous dispersion) phasing is a popular method for solving new macromolecular crystal structures, but it requires that at least two data sets (including Bijvoet pairs) be measured at appropriate wavelengths in the vicinity of the absorption edge of an anomalous scatterer. However, when SIRAS (single isomorphous replacement with anomalous scattering) data are available, “quasi-MAD” data sets can be created based on the SIRAS amplitudes once the heavy-atom substructure has been determined, and then the MAD formalism can be applied. Quasi-MAD phasing is MAD phasing using such quasi-MAD data sets.

Quasi-MAD phasing can also be applied to a SAD (single-wavelength anomalous dispersion) data set by treating it as a special case of SIRAS. An advantage of this approach to the SAD case is that the proper enantiomorph can be readily determined.

Quasi-MAD phasing has been applied to methylmalonyl-CoA epimerase (PDB accession code 1JC4) [1] using peak-wavelength SAD data. The locations of 22 of the 24 selenium sites were determined, and then substructure phase refinement and solvent flattening were carried out by program BnP [2]. These results will be compared to results obtained from other methods.

This research was supported by NIH grant GM-46733.