Molecular replacement with NMR models using distance-derived pseudo B factors

Area of Science:

• Structural Biology
• Biophysics
• Computational Crystallography

Background:

• Molecular replacement (MR) is a key technique for solving crystal structures.
• NMR models offer valuable structural information but can be challenging templates for MR.
• Standard MR protocols often struggle with the flexibility and inherent errors in NMR-derived models.

Purpose of the Study:

• To investigate the impact of distance-derived B factors on the success of MR using NMR models.
• To enhance the statistical significance and detectability of MR solutions.
• To evaluate the effectiveness of NMR-derived models as templates in crystallographic structure determination.

Main Methods:

• Utilizing pseudo B factors derived from atomic distances within an ensemble of NMR models and an averaged model.
• Applying these modified models to a test case: a Pleckstrin homology domain:ligand complex X-ray structure.
• Assessing MR solution correctness (rotational/translational accuracy) and statistical significance (R factors, correlation coefficients).

Main Results:

• Uniform B factors failed to yield detectable MR solutions.
• Models incorporating distance-derived B factors produced statistically significant R factors and correlation coefficients.
• Distance-derived B factors, especially those comparing NMR models to the X-ray structure, improved MR solution detectability.

Conclusions:

• Distance-derived B factors are crucial for enhancing the statistical significance of MR solutions when using NMR models.
• This approach significantly improves the utility of NMR ensembles as templates for solving crystal structures.
• The method holds promise for future MR problems involving NMR structural data.