Scotty: lattice coincidences for macromolecular crystallographic phasing

Area of Science:

• Crystallography
• Structural Biology
• Computational Biology

Background:

• Macromolecular crystallography is crucial for determining 3D structures of biological molecules.
• Phasing is a critical step in solving crystal structures, often relying on molecular replacement.
• Identifying similar crystal lattices can significantly streamline the phasing process.

Purpose of the Study:

• To introduce Scotty, a software tool for identifying coincident crystal lattices.
• To provide an alternative phasing method to molecular replacement when lattices are closely related.
• To enable efficient structure determination for challenging crystallographic datasets.

Main Methods:

• Utilizing the Phasertng codebase and a scikit-learn `BallTree' index with the Niggli cell distance metric (`NCDist').
• Implementing a nearest-neighbor search to identify potential coincident lattices.
• Correlating structure-factor intensities to select the best template for phasing and refinement.

Main Results:

• Scotty successfully identifies coincident lattices, even with significant non-isomorphism.
• The method accurately phases target lattices using highly correlated structures.
• The nearest-neighbor search is space-group agnostic, allowing for flexible lattice matching.

Conclusions:

• Scotty offers a robust and efficient method for phasing macromolecular crystals with related lattices.
• This approach enhances the speed and accuracy of structure determination in crystallography.
• The software facilitates phasing even when space groups differ or exhibit subgroup relationships.