RestraintLib

RestraintLib provides updated stereochemical restraints for nucleic acids to improve geometric accuracy during macromolecular structure refinement.

Key Features:

• Updated Stereochemical Restraints: Restraints are derived from high-resolution crystal structures in the Cambridge Structural Database and tailored to nucleic acids for use in refinement of lower-resolution experimental models.
• Conformation-Dependent Constraints: Implements conformation-dependent covalent geometry for nucleic acid sugars that varies with ribose type, attached nucleobase, ring puckering, and torsion angles including glycosidic (χ) and side-chain (γ).
• Functional Dependencies: Encodes dependencies of glycosidic-link geometry on χ and of endocyclic ribose bond angles on sugar pucker amplitude (τm).
• Validation and Testing: Restraints have been validated against the Nucleic Acid Database and an ultrahigh-resolution Z-DNA structure in the Protein Data Bank and tested by re-refinement of hundreds of PDB crystal structures.
• Output Formats for Refinement Programs: Provides restraint representations compatible with REFMAC, PHENIX, and SHELXL for integration into refinement workflows.

Scientific Applications:

• Macromolecular structure refinement: Improves geometric restraints used during crystallographic and cryo-EM refinement of nucleic acid-containing models.
• Structural biology and drug design: Enables more accurate nucleic acid geometries that inform structure-based drug design and interpretation of ligand interactions.
• Molecular genetics: Supports more precise modeling of nucleic acid conformations relevant to sequence-dependent structural studies.
• Bioinformatics: Provides standardized restraint sets for computational analyses and comparative refinement studies across PDB entries.

Methodology:

Restraints were derived from analysis of high-resolution crystal structures in the Cambridge Structural Database, implemented as conformation-dependent covalent geometry varying with ribose type, nucleobase, ring pucker, χ and γ torsions, and with glycosidic and endocyclic bond-angle dependencies on χ and τm; validation used the Nucleic Acid Database and an ultrahigh-resolution Z-DNA PDB entry and testing involved re-refinement of hundreds of PDB crystal structures.