AWSEM-Suite: a protein structure prediction server based on template-guided, coevolutionary-enhanced optimized...

Area of Science:

• Computational biology
• Structural biology
• Bioinformatics

Background:

• Accurate protein 3D structure prediction remains challenging despite advances in experimental and computational methods.
• Predicting protein tertiary structures is crucial for understanding protein function and biological processes.

Purpose of the Study:

• To introduce AWSEM-Suite, a free and open-access web server for predicting monomeric protein tertiary structures from amino acid sequences.
• To enhance protein structure prediction accuracy by integrating homologous templates and co-evolutionary data.

Main Methods:

• Development of AWSEM-Suite based on a coarse-grained protein force field optimized using neural network principles and energy landscape theory.
• Incorporation of physically motivated potentials, knowledge-based local structure biasing, homologous template restraints, and co-evolutionary restraints.
• Evaluation using independent metrics from the CASP13 (Critical Assessment of protein Structure Prediction) experiment.

Main Results:

• AWSEM-Suite demonstrated improved predictive power compared to pure AWSEM structure prediction.
• The server achieved the eighth position in the free modeling category of CASP13, indicating reasonable accuracy.
• The web server provides a user-friendly interface for accessing its prediction capabilities.

Conclusions:

• AWSEM-Suite is a powerful and accessible tool for predicting monomeric protein tertiary structures.
• It is particularly valuable when no suitable structure template is available.
• The integration of various restraints significantly enhances prediction accuracy.