A method for evaluating the structural quality of protein models by using higher-order phi-psi pairs scoring

Area of Science:

• Structural biology
• Computational biology
• Biophysics

Background:

• Accurate protein structure determination is crucial for understanding biological function.
• Evaluating the quality of experimentally or computationally derived protein models remains a challenge.
• Existing methods may not fully capture the conformational preferences of protein fragments.

Purpose of the Study:

• To develop and validate a novel method for assessing the quality of protein structural models.
• To utilize phi-psi angle distributions within protein fragments as a measure of structural plausibility.
• To establish a quantitative criterion for distinguishing high-quality protein models.

Main Methods:

• A sliding window approach was employed to dissect protein models into small fragments (3-12 residues).
• Each fragment was represented by a vector of phi-psi angles.
• Fragment conformations were scored based on their fit to a database of phi-psi angles from high-resolution X-ray crystal structures.
• The method focuses on the conformational space of longer fragments as a key discriminator.

Main Results:

• The proposed method effectively scores the quality of both experimental and theoretical protein structures.
• Measuring the fit of predicted models to the allowed conformational space of longer fragments significantly discriminates model quality.
• Protein models with higher-order phi-psi score fit values (m) greater than -1.00 were deemed reasonable.

Conclusions:

• The phi-psi angle-based fragment scoring method provides a robust metric for protein model quality assessment.
• This approach enhances the evaluation of structural models by considering local conformational preferences within a broader context.
• The findings offer a valuable tool for researchers in structural biology and drug discovery.