Phasing via pure crystallographic least squares: an unexpected feature

Area of Science:

• Crystallography
• Structural Biology
• Computational Chemistry

Background:

• Crystallographic least-squares techniques are standard for refining known crystal structures.
• Ab initio phasing, determining crystal structures from scratch, traditionally requires different methods.
• Convergence limitations of least-squares methods have hindered their use in ab initio phasing.

Purpose of the Study:

• To explore the application of crystallographic least-squares techniques for ab initio phasing.
• To develop a method overcoming convergence limits for direct structure determination.
• To assess the feasibility of solving small molecule structures without auxiliary tools like electron-density maps.

Main Methods:

• Utilizing crystallographic least-squares techniques in a novel multi-solution approach.
• Implementing procedures to identify and remove incorrectly positioned atoms.
• Incorporating correctly positioned atoms into an evolving structural model.
• Directly solving crystal structures without calculating electron-density maps.

Main Results:

• Demonstrated the successful application of least-squares techniques for ab initio phasing.
• Overcame the convergence limitations previously associated with this method.
• Solved numerous small molecular structures efficiently.
• Showcased structure determination without reliance on electron-density maps.

Conclusions:

• Least-squares techniques can be effectively employed for ab initio phasing.
• The developed multi-solution procedure offers a robust solution to convergence challenges.
• This approach provides a powerful, streamlined method for solving small crystal structures.