Localization of Millisecond Dynamics: Dihedral Entropy from Accelerated MD

Area of Science:

• Computational Biology
• Biophysics
• Molecular Dynamics

Background:

• Understanding protein dynamics is crucial for molecular biology.
• Current simulation methods are limited in capturing long-timescale dynamics.
• Local protein flexibility influences protein function and interactions.

Purpose of the Study:

• To develop and validate a novel computational method for enhanced sampling of local protein dynamics.
• To extend the accessible timescale for molecular dynamics simulations by three orders of magnitude.
• To provide residue-wise insights into protein flexibility on biologically relevant timescales.

Main Methods:

• Accelerated molecular dynamics (aMD) simulations were employed for conformational sampling.
• Reweighted backbone dihedral distributions were extracted from simulation data.
• Torsional probabilities were used to characterize local dynamics and calculate dihedral entropies.
• The method was validated on alanine dipeptide, BPTI, and Bet v 1a protein systems.

Main Results:

• The method successfully captured local protein dynamics on timescales significantly beyond conventional simulations.
• Calculated flexibility profiles showed excellent agreement with large-scale computer simulations.
• The results were further validated by comparison with Nuclear Magnetic Resonance (NMR) experimental data.
• Residue-wise dihedral entropies provided detailed insights into protein flexibility.

Conclusions:

• The developed method offers an efficient approach to study local protein dynamics.
• This technique significantly extends the accessible timescales for molecular dynamics simulations.
• The findings provide valuable insights into protein flexibility relevant to biological function.