Systematically Improvable and Locality Accelerated Enzymatic Reactivity Modeling: Toward Chemical Accuracy at...

Area of Science:

• Computational chemistry
• Biochemistry
• Enzyme kinetics

Background:

• Quantum mechanics/molecular mechanics (QM/MM) is crucial for computational enzymology.
• A major challenge is achieving accurate QM methods for large QM subsystems affordably.

Purpose of the Study:

• To develop a cost-effective QM/MM scheme for accurate simulations of large enzyme systems.
• To establish guidelines for implementing this new computational protocol.

Main Methods:

• Utilizing local natural orbital (LNO)-based CCSD(T) for accurate energetics and reference.
• Selecting reliable hybrid density functional theory (DFT) models for large QM subsystems.
• Employing quantum embedding, accelerated by the local embedded subsystem (LESS) approach, to reduce DFT costs.

Main Results:

• The proposed Locality Accelerated and Systematically Improvable (LASI) scheme allows for large QM subsystems (up to ~400 atoms) with DFT accuracy.
• The LASI protocol significantly reduces computational cost, making complex QM/MM simulations feasible.
• Validation through enzyme-catalyzed phosphate hydrolysis demonstrates the protocol's effectiveness for clinically relevant reactions.

Conclusions:

• The LASI scheme offers a flexible and affordable method for QM/MM simulations, balancing QM accuracy with computational efficiency.
• This approach is broadly applicable for predictive computational studies of enzyme reactivity.
• The study provides general guidelines for setting up LASI protocol components for diverse QM/MM applications.