A genetic algorithm for the ab initio phasing of icosahedral viruses

Area of Science:

• Structural Biology
• Computational Biology
• Biophysics

Background:

• De novo phasing of low-resolution X-ray diffraction data is crucial for determining the structure of icosahedral viruses.
• Existing methods may face limitations with limited prior knowledge of viral shape and size.

Purpose of the Study:

• To investigate the application of genetic algorithms for de novo phasing of low-resolution X-ray diffraction data from icosahedral viruses.
• To develop an efficient computational method for initial structure determination without extensive prior information.

Main Methods:

• Modeling the virus as a symmetry expansion of lattice points to sample the icosahedrally unique volume.
• Utilizing a matrix formulation for efficient model evaluation and survey of possible models.
• Refining candidate solutions and selecting trials based on intensity-based statistics across all resolution ranges.

Main Results:

• Genetic algorithms provide a parameterization for efficient model evaluation and survey of potential virus structures.
• An initial model can be generated with minimal prior information, yielding a reasonable low-resolution image.
• This approach successfully provides an acceptable starting point for symmetry-based direct phase extension approximately 50% of the time.

Conclusions:

• Genetic algorithms are effective computational tools for the de novo phasing of low-resolution X-ray diffraction data from icosahedral viruses.
• The described parameterization and refinement strategy offer an efficient pathway to initial viral structure determination.
• Further improvements in efficiency can be achieved by integrating selection criteria directly into the genetic algorithm's fitness function.