Strategy for supplementing structure calculations using limited data with hydrophobic distance restraints

Area of Science:

• Biochemistry
• Structural Biology
• Computational Biology

Background:

• Protein structure is crucial for function.
• Dihedral angles (phi and psi) are traditionally considered sufficient to define protein fold.
• This study investigates the sufficiency of dihedral angle restraints for accurate protein structure prediction.

Purpose of the Study:

• To test the hypothesis that dihedral angles alone define protein fold.
• To evaluate the effectiveness of Torsion Angle Dynamics/Simulated Annealing (TAD/SA) with dihedral restraints.
• To develop and test improved TAD/SA methods incorporating long-range restraints.

Main Methods:

• Torsion Angle Dynamics and Simulated Annealing (TAD/SA) calculations.
• Utilized dihedral angle restraints based on X-ray crystallography data.
• Incorporated hydrogen-bond restraints and hydrophobic distance restraints.
• Tested with simulated restraints from various protein structures and NMR data.

Main Results:

• Dihedral angle restraints alone resulted in significant deviation (4 Å RMSD) from the native protein structure.
• Deviations were attributed to non-planar peptide bonds, energy conflicts, and lack of long-range information.
• Including even a few long-range contacts (hydrogen bonds) dramatically improved fold accuracy.
• The developed TAD/SA procedure with combined restraints showed improved performance.

Conclusions:

• Dihedral angles alone are insufficient to accurately predict protein structure.
• Long-range interactions, particularly hydrogen bonds, are critical for defining protein fold.
• A combined restraint strategy improves the accuracy of computational protein structure prediction.