Homology-based hydrogen bond information improves crystallographic structures in the PDB

Area of Science:

• Structural Biology
• Biochemistry
• Computational Biology

Background:

• The Protein Data Bank (PDB) archives macromolecular structures, serving over a million users annually.
• Crystallographic refinement in the PDB traditionally overlooks hydrogen bond (H-bond) information.
• H-bond restraints can significantly improve low-resolution structures where diffraction data is limited.

Purpose of the Study:

• To develop and implement methods for deriving H-bond information to enhance low-resolution crystallographic structure refinement.
• To create an updated PDB-REDO databank incorporating these new H-bond restraints for public access.

Main Methods:

• Deriving H-bond information globally from high-resolution structures in PDB-REDO.
• Deriving H-bond information specifically from homologous high-resolution structures.
• Implementing refinement using HOmology DErived Restraints (HODER) and applying it to all PDB crystallographic structures using parallel computing.

Main Results:

• HODER refinement improves the geometrical quality of low-resolution structures.
• The refined structures exhibit a better fit to the experimental diffraction data.
• A comprehensive, updated PDB-REDO databank has been generated with these improvements.

Conclusions:

• The new HODER method effectively leverages H-bond information to enhance crystallographic structure refinement, particularly at low resolutions.
• The updated PDB-REDO databank provides a valuable, uniformly treated resource for researchers studying individual structures, protein families, and general protein structural features.