Macromolecular X-ray structure determination using weak, single-wavelength anomalous data

Abstract

A likelihood-based method for determining the sub-structure of anomalously-scattering atoms in macromolecular crystals can allow successful structure determination by single-wavelength anomalous diffraction (SAD) X-ray analysis with weak anomalous signal. Along with use of partial models and electron density maps in searches for anomalously-scattering atoms, testing of alternative values of parameters, and parallelized automated model-building, this method has the potential for extending the applicability of the SAD method in challenging cases.

Selenium single-wavelength anomalous diffraction de novo phasing using an X-ray-free electron laser

“…Supplementary Fig. 2 shows a plot of the correlation of the final maps with the known structure and anomalous peak/root-mean-square (RMS)30 ratio as a function of the percentage of the total data used.…”

Maximum-likelihood determination of anomalous substructures

“…Several test cases were used during development to establish sensible defaults and to gauge the performance of the new algorithm, which was benchmarked against the current HySS algorithm that includes Phaser log-likelihood-gradient completion (Bunkó czi et al, 2015). The cases were chosen to sample substructures with different levels of anomalous signal and to evaluate the effect of accounting for the real scattering contribution of the anomalous scatterers.…”

“…It is possible to solve this structure by SAD phasing using the data from either wavelength, but it is much more difficult with the high-energy remote data set (wavelength of 0.91837 Å ) used in the tests reported here. A substructure containing all seven Se sites can be determined with HySS in 1171 s when the Phaser log-likelihood-gradient completion algorithm is used, but not when HySS is confined research papers to the earlier direct-methods approaches (Bunkó czi et al, 2015).…”

“…A significant component of the success of MLSAD has been the use of log-likelihood-gradient maps (McCoy & Read, 2010;Read & McCoy, 2011), rather than anomalous difference Fouriers, to edit and complete the substructures, an approach introduced in SHARP (de La Fortelle & Bricogne, 1997). Recently, it was shown that substructure determination could be significantly strengthened by giving a prominent role to MLSAD throughout the process, rather than just in adding weak sites to the bulk of the substructure (Bunkó czi et al, 2015); HySS was modified so that MLSAD log-likelihood-gradient substructure completion took over after as few as two sites had been determined.…”

Improving the efficiency of molecular replacement by utilizing a new iterative transform phasing algorithm

“…To assess how maps from the HIO iteration compare with those obtainable from several other common density-modification schemes, we ran trials using Solomon (Abrahams & Leslie, 1996), DM (Cowtan, 1994), the prime-and-switch method (Terwilliger, 2004) using phenix.-autobuild and phenix.resolve, Pirate (Cowtan, 2000) and Parrot (Cowtan, 2010). As pointed out by Bunkó czi et al (2015), such comparisons are inherently biased as we are better at optimizing our own code than that of others. We sought to mitigate this bias by testing runs with multiple solvent contents, starting maps and program parameters around the defaults, but ultimately our results are a lower bound for the phase improvement from these methods.…”