CNSsolve

CNSsolve performs macromolecular structure determination using X-ray crystallography and solution nuclear magnetic resonance (NMR) spectroscopy and supports electron microscopy and solid-state NMR methods.

Key Features:

• Flexible Architecture: A hierarchical architecture exposes layered interfaces including task-oriented input files, module files, and the symbolic CNS language to access internal data structures.
• Extensible Framework: The CNS language enables implementation of new algorithms and extension to additional structure-determination methods without modifying core source code.
• Comprehensive Procedures: Supports operations on structure factors, electron-density maps, and atomic properties and provides crystallographic procedures for phasing, density modification, and refinement.
• Modular Design and Disordered Solvent Modeling: Modular framework includes a disordered solvent model that combines grid search with least-squares optimization to improve robustness at lower resolution and reduce R values.
• Thermal Factor Sharpening: Applies thermal factor sharpening to electron density maps to enhance map interpretability.
• Integration of Information Sources: Integrates multiple sources of experimental and derived information into stages of structure determination for combined analysis of structural and dynamical properties.

Scientific Applications:

• X-ray Crystallography: Phasing, density modification, and refinement of crystallographic data including treatment of disordered solvent and map sharpening.
• Solution NMR Spectroscopy: Macromolecular structure determination and analysis of dynamics from solution NMR data.
• Electron Microscopy and Solid-State NMR: Support for alternative structure-determination methods beyond crystallography and solution NMR.
• Low-Resolution Model Refinement: Refinement strategies and solvent modeling to improve model quality and R values at lower resolution.
• Integrated Structural and Dynamical Analysis: Combined analysis of structural and dynamical properties by integrating diverse data sources.

Methodology:

Computational methods include a hierarchical architecture and the symbolic CNS language; operations on structure factors, electron-density maps, and atomic properties; crystallographic procedures for phasing, density modification, and refinement; a disordered solvent model combining grid search with least-squares optimization; thermal factor sharpening of maps; and integration of multiple information sources.