SOESA

SOESA refines atomic models of protein structures by integrating a database-derived interatomic distance probability density function into a database potential used within molecular dynamics to improve model accuracy, particularly for low-resolution X-ray crystal data.

Key Features:

• Interatomic Distance Probability Density Function (PDF) Database: Utilizes a comprehensive database compiled from known protein structures to derive an interatomic distance PDF for estimating prior probabilities of interatomic distances.
• Database Potential Energy Function: Employs a database potential energy function derived from interatomic distance distributions that replaces traditional Van der Waals potentials in simulations and structural refinements.
• Molecular Dynamics Methodology: Integrates the database potential into molecular dynamics to refine atomic positions, interatomic distances, and side-chain dihedral angles, improving placement of characteristically buried groups.
• Refinement with Low-Resolution Data: Demonstrates efficacy in refining X-ray crystal structures truncated at low resolution (e.g., ~3.5 Å), yielding improved R-factors and more accurate calculated model phases.

Scientific Applications:

• Low-resolution X-ray structure refinement: Improves atomic detail and model accuracy for X-ray crystal structures when high-resolution data are unavailable.
• Initial model optimization: Enhances structure determination from randomized or poorly defined starting models during early refinement stages.
• Side-chain and buried-group placement: Provides improved positioning of side chains and buried chemical groups compared with standard refinement techniques.
• R-factor and phase improvement: Produces improved R-factor analyses and more accurate calculated model phases, contributing to more precise atomic arrangements.

Methodology:

Refinement of X-ray crystal structures using molecular dynamics simulations that incorporate the database-derived interatomic distance PDF as a database potential energy function to optimize atomic positions, interatomic distances, and side-chain dihedral angles, with notable improvements for data truncated to ~3.5 Å.