Improved modeling of side-chains in proteins with rotamer-based methods: a flexible rotamer model

Area of Science:

• Computational Biology
• Structural Bioinformatics
• Protein Science

Background:

• Side-chain modeling is crucial for protein structure determination, prediction, and design.
• Rotamer-based methods are efficient but traditionally use rigid conformations.
• A need exists for more accurate and flexible side-chain modeling approaches.

Purpose of the Study:

• To develop and validate a flexible rotamer model for side-chain modeling in proteins.
• To thermodynamically calculate effective energies for flexible rotamers.
• To compare the performance of the flexible rotamer model against the rigid rotamer model and other state-of-the-art methods.

Main Methods:

• Developed a flexible rotamer model representing ensembles of conformations.
• Utilized a thermodynamically based method to compute effective energies for flexible rotamers.
• Applied the flexible and rigid rotamer models within a self-consistent mean field theory method to predict protein structures.

Main Results:

• The flexible rotamer model achieved improved prediction accuracy: 85.8% for chi1, 76.5% for chi1+2, and 1.34 Å RMSD.
• Core residues showed even greater improvement with the flexible model: 93.0% for chi1, 87.7% for chi1+2, and 0.70 Å RMSD.
• Flexible rotamer model predictions significantly outperformed the rigid model and another leading method, especially for core residues.

Conclusions:

• The flexible rotamer model offers a significant advancement over the classic rigid rotamer model.
• This model can be advantageously integrated into various rotamer-based methods for protein structure determination, prediction, and design.
• The flexible rotamer approach also shows promise for predicting free energies in mutational studies.