KoBaMIN

KoBaMIN minimizes a knowledge-based potential of mean force (PMF) derived from the Protein Data Bank (PDB) to refine protein structures and improve model stereochemistry and energetic plausibility.

Key Features:

• Knowledge-based PMF: Minimizes a potential of mean force derived from extensive structural data in the Protein Data Bank (PDB), with the energy function implicitly accounting for solvent and crystal environment influences.
• Input handling: Refines both single protein structures and ensembles starting from their coordinates in PDB format.
• Stereochemistry optimization: Offers an optional MESHI-based stereochemistry optimization step to improve stereochemical quality.
• Structure comparison: Includes functionality to compare refined structures with a reference structure to evaluate refinement-induced changes.
• Computational performance: Demonstrates computational efficiency and accuracy enabling rapid and precise refinements.
• Benchmark results: Produced robust performance on CASP7 decoys and achieved top-ranking refinements in CASP8 and CASP9.

Scientific Applications:

• Model refinement: Improve stereochemistry and energetic plausibility of predicted protein models.
• Ensemble refinement: Refine ensembles of structural models derived from PDB coordinates.
• Method benchmarking: Assess refinement protocols and compare performance using decoy sets such as CASP7 decoys and CASP competitions.
• Structure validation: Quantify refinement effects by comparison to reference structures.
• Structural biology studies: Enhance structural accuracy to support investigations of protein function and interactions.

Methodology:

Minimization of a knowledge-based PMF derived from PDB structural data; optional stereochemistry optimization using MESHI; comparison of refined models to reference structures.