Accelerating Quantum Mechanical/Molecular Mechanical (QM/MM) Simulations with Stochastic Iso-Kinetic Nosé-Hoover...

Area of Science:

• Computational Chemistry
• Biochemistry
• Molecular Dynamics

Background:

• Ab initio quantum mechanical/molecular mechanical (ai-QM/MM) simulations are computationally expensive, limiting their application in large-scale enzymatic mechanism studies.
• High computational cost is a major barrier to widespread use of ai-QM/MM for complex modeling.

Purpose of the Study:

• To develop and validate an accelerated ai-QM/MM simulation method to overcome computational cost limitations.
• To enable efficient large-scale modeling of enzyme mechanisms.

Main Methods:

• Implementation of a stochastic iso-kinetic Nosé-Hoover chain (SIN(R)) thermostat with a multiple time step (MTS) integration scheme in CHARMM.
• Periodic application of ai-QM/MM corrections and use of augmented molecular mechanics (MM) bonded terms in the quantum mechanics (QM) region.
• Testing on S_N2 reactions, dihydrofolate reductase (DHFR) hydride transfer, and adenylate kinase (AK) phosphoryl transfer.

Main Results:

• The SIN(R)-based MTS method achieved stable simulations with outer time steps up to 10 fs for S_N2 reactions, with minimal error in barrier height (<1.0 kcal/mol).
• Stable dynamics for DHFR were achieved with 10 fs outer steps.
• A 4-fold speed-up was observed for semiempirical QM/MM simulations of adenylate kinase using a 2 fs outer time step compared to conventional 0.5 fs simulations.

Conclusions:

• The SIN(R)-based MTS method substantially accelerates QM/MM simulations, making it a practical tool for enzymatic catalysis mechanism studies.
• This approach effectively reduces the number of costly ai-QM/MM calculations while maintaining accuracy.