A fragmentation and reassembly method for ab initio phasing.

Area of Science:

• Structural biology
• X-ray crystallography
• Computational biology

Background:

• Ab initio phasing using de novo models is a key technique for solving protein structures from crystallographic data.
• Current de novo modeling methods often produce models with insufficient accuracy for direct use in molecular replacement.

Purpose of the Study:

• To develop and evaluate a novel fragment-assembly phasing method to improve ab initio structure determination.
• To enable successful molecular replacement with low-accuracy de novo models.

Main Methods:

• A fragment-assembly phasing approach was developed, disassembling low-accuracy de novo models into fragments.
• Fragments were independently placed via molecular replacement and reassembled into a complete structure.
• Automated model building was used for final structure determination.

Main Results:

• The fragment-assembly method successfully solved structures for 8 out of 10 protein targets.
• The best de novo models were, on average, over 4.0 Å from the native structures, proving insufficient for direct molecular replacement.
• The method extends the applicability of ab initio phasing when de novo models have limited accuracy.

Conclusions:

• Fragment-assembly phasing is a viable strategy for ab initio structure determination when direct de novo models fail.
• This method enhances the utility of computational modeling in protein crystallography.
• It expands the scope of solving protein structures from crystallographic data.