Fast protein loop sampling and structure prediction using distance-guided sequential chain-growth Monte Carlo method.

Area of Science:

• Computational biology
• Structural biology
• Biophysics

Background:

• Protein loops are flexible regions connecting secondary structures.
• Their flexibility makes experimental determination and computational modeling difficult.
• Accurate loop modeling is crucial for understanding protein function.

Purpose of the Study:

• To develop an efficient and accurate computational method for protein loop conformation sampling and prediction.
• To improve the modeling of flexible loop regions in proteins.

Main Methods:

• Developed Distance-guided Sequential chain-Growth Monte Carlo (DISGRO) method.
• Utilized a chain growth sequential Monte Carlo sampling strategy.
• Employed a specialized energy function for loops and a novel geometric criterion.

Main Results:

• DISGRO efficiently generates high-quality, low-energy loop conformations enriched with near-native structures.
• Achieved average minimum global backbone RMSD of 1.53 Å for 12-residue loops.
• Demonstrated significant reduction in computational cost compared to existing methods (e.g., FALCm).
• Showed comparable or superior performance to previous methods, especially for longer loops (10-17 residues).

Conclusions:

• DISGRO provides an efficient and effective approach for protein loop modeling.
• The method significantly reduces computational time while maintaining high accuracy.
• DISGRO is particularly advantageous for modeling longer and more challenging protein loops.