Related Experiment Videos
Recognizing misfolded and distorted protein structures by the assumption-based similarity score
A P Golovanov1, P E Volynsky, S B Ermakova
1Shemyakin and Ovchinnikov Institute of Bioorganic Chemistry, Russian Academy of Sciences, Moscow.
Protein Engineering
|March 5, 1999
Summary
A novel sigma-score accurately identifies correct protein structures and distinguishes them from misfolded models. This new scoring method favors native-like contacts, improving protein structure evaluation and selection.
Area of Science:
- Computational Biology
- Structural Biology
- Biophysics
Background:
- Accurate protein structure determination is crucial for understanding biological function.
- Distinguishing correct protein models from subtle alternatives and misfolded structures remains a challenge.
- Existing methods may struggle with minor structural deviations.
Purpose of the Study:
- To introduce a new similarity score, the sigma-score, for evaluating protein structures.
- To assess the capability of the sigma-score in identifying correct protein conformations.
- To differentiate between native and misfolded protein structures, even with minimal deviations.
Main Methods:
- Development of the sigma-score based on the principle 'similar likes similar'.
- Favoring hydrophobic/hydrophilic contacts and disfavoring mixed contacts.
- Testing the sigma-score against experimental structures and various alternatives (minimized, molecular dynamics in vacuo/solvent, misfolded models).
- Estimating confidence intervals using non-crystallographic symmetry in X-ray structures.
Main Results:
- The sigma-score successfully distinguished correct protein structures from close alternatives in most cases.
- Effective differentiation was achieved even when root mean square displacement was less than 1 Å.
- The score demonstrated robustness across various types of alternative structures and simulation conditions.
Conclusions:
- The sigma-score is a reliable metric for assessing overall protein model quality.
- It effectively selects correct structures from closely related alternatives.
- The score aids in characterizing protein folding/unfolding simulations and evaluating structural models.