Application of the minimal principle to peptide structures

Area of Science:

• Crystallography
• Structural Biology
• Computational Chemistry

Background:

• Direct methods are crucial for solving crystal structures.
• Accurate phase determination is a key challenge in crystallography.
• Automating structure solution is highly desirable for efficiency.

Purpose of the Study:

• To develop a new, automated direct-methods procedure for crystal structure solution.
• To refine crystal structure phasing through iterative Fourier summation and real space filtering.
• To assess the potential of this method for solving medium-to-large sized structures.

Main Methods:

• A novel direct-methods procedure involving phase refinement via the minimal function R(phi).
• Alternating phase refinement with Fourier summation and real space filtering.
• Initial phase assignment using randomly positioned atoms and structure factors.
• Minimizing R(phi) using a parameter shift method.
• Iterative refinement using electron density peaks as trial structures.

Main Results:

• The devised procedure combines phase refinement, Fourier summation, and real space filtering.
• Initial random phases are refined by minimizing the R(phi) function.
• Refined phases lead to electron density maps from which new trial structures are generated.
• Convergence to a solution depends on structural complexity and refinement parameters.
• The method shows potential for automated structure solutions in the 200-400 atom range.

Conclusions:

• A new direct-methods procedure has been successfully developed.
• The iterative approach of phase refinement and real space filtering is effective.
• This automated method holds significant promise for routine crystal structure solution.
• The procedure is particularly suited for structures containing 200-400 atoms.