Deriving a Polarizable Force Field for Biomolecular Building Blocks with Minimal Empirical Calibration.

Area of Science:

• Computational Chemistry
• Molecular Modeling
• Biophysics

Background:

• Force field parametrization is crucial for molecular simulations but traditionally relies on laborious experimental fitting.
• Existing methods face challenges due to complex, underdetermined optimization problems and time-consuming calibration.

Purpose of the Study:

• To develop and validate a method for deriving force field parameters directly from electronic structure calculations.
• To investigate a polarizable force field model incorporating higher-order dispersion terms for biomolecular simulations.

Main Methods:

• Utilized quantum mechanical/molecular mechanical (QM/MM) calculations to derive atomic polarizabilities, partial charges, and higher-order dispersion terms.
• Applied these quantum-mechanically determined parameters to a set of 49 biochemically relevant molecules.
• Focused on updating parameters for hydrocarbon side chains.

Main Results:

• Successfully derived and applied QM/MM parameters for a polarizable model with higher-order dispersion.
• Demonstrated that quantum-mechanically determined parameters can fix a large majority (132 out of 138) of nonbonded parameters.
• Presented updated parameters for hydrocarbon side chains based on calculations for 49 molecules.

Conclusions:

• Quantum-mechanically derived parameters significantly streamline the force field parametrization process.
• The proposed method effectively reduces the parameter space and labor associated with traditional fitting approaches.
• This approach enhances the accuracy and efficiency of biomolecular simulations by leveraging electronic structure data.