Multivariate phase combination improves automated crystallographic model building.

Area of Science:

• Macromolecular crystallography
• Structural biology
• Biophysics

Background:

• Density modification is crucial for improving electron-density maps in macromolecular crystallography.
• Current methods assume independence between initial and density-modified maps, indirectly propagating map accuracy.
• This indirect propagation can limit the effectiveness of density modification.

Purpose of the Study:

• To develop a novel multivariate equation for density modification that does not assume map independence.
• To directly incorporate observed diffraction data and refine errors within the density modification process.
• To significantly improve the quality of electron-density maps and facilitate automated structure building.

Main Methods:

• Derivation of a multivariate equation integrating initial and density-modified maps without assuming independence.
• Direct utilization of observed diffraction data and refinement of errors from single-wavelength anomalous diffraction (SAD) experiments.
• Implementation and validation of the new method on over 100 real crystallographic data sets.

Main Results:

• The new method yields significantly improved electron-density maps compared to current state-of-the-art techniques.
• The approach directly refines errors inherent in SAD experiments.
• Demonstrated substantial enhancement in map quality across numerous real-world datasets.

Conclusions:

• The derived multivariate equation represents a breakthrough in density modification for macromolecular crystallography.
• Directly accounting for map dependencies and experimental errors leads to superior electron-density map quality.
• This advancement promises to increase the success rate of automated protein structure determination.