Computational Studies of Intrinsically Disordered Proteins.

Area of Science:

• Biophysics
• Computational Biology
• Protein Science

Background:

• Intrinsically disordered proteins (IDPs) are challenging to study experimentally.
• Molecular dynamics (MD) simulations offer insights but face limitations like force field bias and insufficient sampling.
• Previous computational studies often show discrepancies with experimental data.

Purpose of the Study:

• To assess limitations in intrinsically disordered protein (IDP) simulations, specifically force field performance and sampling sufficiency.
• To compare an IDP-specific force field (ff14IDPSFF) against a generic one (ff14SB) using short peptides and the HIV-1 Rev protein.
• To investigate the structural preferences and conformational dynamics of apo Rev.

Main Methods:

• Simulated nine unstructured short peptides and the HIV-1 Rev protein using molecular dynamics.
• Extended simulations to microseconds to evaluate force field performance and sampling.
• Utilized NMR observables and structural analyses to compare simulated results with experimental data and assess secondary structure preferences.

Main Results:

• The IDP-specific force field ff14IDPSFF showed improved agreement with experimental NMR observables for short peptides compared to the generic ff14SB.
• The advantages of ff14IDPSFF were less clear in simulations of apo Rev.
• Even 10 ms of simulation time with ff14SB for apo Rev suggested insufficient sampling, highlighting the need for enhanced sampling techniques.
• Structural analyses revealed distinct secondary structural preferences: ff14SB favored helical structures, while ff14IDPSFF favored random coil conformations for apo Rev.

Conclusions:

• IDP-specific force fields can enhance the accuracy of molecular dynamics simulations for certain IDPs, particularly short peptides.
• Sufficient sampling remains a critical challenge in IDP simulations, even with extended timescales and specialized force fields.
• Further quantitative experimental and computational analyses are needed to determine the optimal force field balance (not too ordered, not too disordered) for complex IDPs like Rev.