RPF: a quality assessment tool for protein NMR structures

Area of Science:

• Biochemistry
• Structural Biology
• Bioinformatics

Background:

• Protein Nuclear Magnetic Resonance (NMR) spectroscopy is crucial for determining 3D structures.
• Assessing the quality and accuracy of these NMR-derived structures is essential for reliable biological interpretation.
• Existing methods may not fully capture the nuances of structural fit to experimental data.

Purpose of the Study:

• To introduce the RPF web server, a novel tool for the quality assessment of protein NMR structures.
• To provide a quantitative measure of structural accuracy using experimental NMR data.
• To facilitate the refinement of protein structures and associated experimental data.

Main Methods:

• The RPF server analyzes the 'goodness-of-fit' between 3D protein structures and experimental NMR chemical shift data.
• It incorporates unassigned Nuclear Overhauser Effect SpectroscopY (NOESY) data into the quality assessment.
• A Discrimination Power (DP) score is calculated by comparing the fit of query structures to random coil models.

Main Results:

• The RPF server provides a reliable Discrimination Power (DP) score, serving as an accuracy predictor for protein NMR structures.
• Local structure quality measures are visualized on both the 3D molecular structure and the NMR spectra.
• The tool enables iterative refinement of structural models and NOESY peak assignments.

Conclusions:

• The RPF web server offers a robust method for evaluating protein NMR structure quality.
• The DP score and visualization tools aid researchers in identifying and correcting structural inaccuracies.
• This resource supports the advancement of high-resolution protein structure determination using NMR.