An evolutionary computational approach to the phase problem in macromolecular X-ray crystallography

Area of Science:

• Structural biology
• Computational chemistry
• Biophysics

Background:

• Determining macromolecular structures is crucial for understanding biological functions.
• Direct phasing from diffraction data is challenging, especially with limited prior structural information.
• Existing methods often require significant a priori knowledge or are computationally intensive.

Purpose of the Study:

• To demonstrate the efficacy of an evolutionary search algorithm for direct phasing of macromolecules.
• To compute molecular envelopes from diffraction amplitudes without prior structural data.
• To assess the general applicability of this computational approach.

Main Methods:

• Ab initio computation of molecular envelopes using diffraction amplitudes.
• Employing an efficient and inherently parallel evolutionary search algorithm.
• Testing the method on two protein examples with minimal a priori structural information.

Main Results:

• Successfully computed molecular envelopes for two proteins exclusively from diffraction amplitudes.
• Demonstrated that the evolutionary search algorithm can assist in direct phasing.
• Showcased the method's ability to work with limited structural information.

Conclusions:

• The evolutionary computational approach is effective for direct macromolecular phasing.
• This method is broadly applicable across various data resolutions, symmetries, and structural sizes.
• The algorithm offers a powerful tool for structural biology when a priori data is scarce.