Developing a molecular dynamics force field for both folded and disordered protein states.

Area of Science:

• Computational biology
• Biophysics
• Structural biology

Background:

• Molecular dynamics (MD) simulations are crucial for understanding protein structural dynamics.
• Current force fields often struggle to accurately model both folded and intrinsically disordered proteins.
• Accurate modeling is essential for characterizing diverse biological systems.

Purpose of the Study:

• To evaluate the accuracy of existing molecular dynamics force fields for both folded and disordered proteins.
• To develop an improved force field capable of simultaneously describing folded and disordered protein characteristics.
• To expand the applicability of MD simulations to a wider range of biological systems.

Main Methods:

• Selected a benchmark set of 21 folded and disordered protein systems.
• Simulated these systems using six state-of-the-art molecular dynamics force fields.
• Compared simulation results against over 9,000 experimental data points for validation.

Main Results:

• No tested force field accurately described folded proteins, disordered protein dimensions, and secondary structure propensities simultaneously.
• A modified force field, a99SB-disp, was developed by optimizing parameters.
• a99SB-disp achieved excellent agreement with experimental data for disordered proteins while maintaining high accuracy for folded proteins.

Conclusions:

• Existing molecular dynamics force fields have limitations in accurately modeling both folded and disordered proteins.
• The newly developed a99SB-disp force field offers improved accuracy for disordered proteins without compromising folded protein performance.
• a99SB-disp significantly enhances the potential of molecular dynamics simulations for studying a broader spectrum of protein structures and functions.